Aspergillus niger β-glucosidase from previous work was used (13). The fungus was isolated from farm soil, and its genus and species were determined before application. The β-glucosidase was purified in two stages of ion-exchange chromatography. Onion was purchased from a local market (Elaheyeh, Iran). Olive leaves were separated from olive trees in local orchards in Kermanshah province, Iran. Quercetin, quercetin-3-O-glucoside, Trolox and 1,1-diphenyl-2-picrylhydrazyl (DPPH) were purchased from Sigma-Aldrich, Merck (St. Louis, MO, USA). Oleuropein was supplied from Cayman Chemical Company (Ann Arbor, MI, USA). Other chemical materials were of analytical grade and obtained from Merck (Darmstadt, Germany).

Onion peel extract was prepared according to Lee et al. (14) with modifications. At first, aqueous ethanol (20 mL of 60% solution) was added to 1 g of dried and powdered red onion peel, and the mixture was stirred at room temperature for 60 min. Then, the mixture was filtered through a Büchner funnel with Whatman No. 2 filter paper, and again 20 mL of solvent were added to the residue. Extraction was repeated for another 30 min. After filtration, the solvent was removed at 40 °C by a rotary evaporator (SB-1200; EYELA, Tokyo, Japan) under a vacuum.

Olive leaf extract was prepared according to Yateem et al. (15) with modifications. Primarily, olive leaves were washed, dried in an oven at 50 °C and powdered. Then, 100 mL of 80% aqueous ethanol were added to 10 g of olive leaf powder, and the mixture was stirred for 4 h at room temperature. After filtration through the Whatman filter paper No. 1, ethanol was evaporated at 40 °C by a rotary evaporator (SB-1200; EYELA) under a vacuum. Then, 15 mL of water was added to the extract, and it was centrifuged (2-16PK; Sigma-Aldrich, Merck, Berlin, Germany) at 4724×g for 5 min to separate chlorophyll and other water-insoluble compounds. Next, 2 mL of dichloromethane were added to the extract, and after mixing for 1 min, the mixture was centrifuged. Finally, the water was evaporated as described in the previous section and the extracts were stored at 4 °C until analysis.

Total phenols were measured before and after the reaction of β-glucosidase with the extracts using the Folin-Ciocalteu reagent according to Roesler et al. (18) with modifications. For this purpose, β-glucosidase was added at concentrations of 1.5 and 2 μg/mL, respectively, in a total volume of 500 µL, to 400 µL of olive leaf (1% m/V) and onion peel (0.5% m/V) extracts. The reaction time of the extracts with β-glucosidase was 24 h at 50 °C for olive leaves and 3 h at 45 °C for onion peel. Enzyme was inactivated by heating the extracts at 90 °C for 5 min. The extracts (100 µL) were mixed with Folin-Ciocalteu reagent (500 µL of 1:10 dilution), and the mixtures were incubated at 25 °C for 5 min. Then, 400 µL of 5% Na₂CO₃ were added, and again incubated in a water bath (ONE-29; Memmert, Schwabach, Germany) at 50 °C for 5 min. Next, the absorbance was read at 765 nm after cooling the samples to room temperature. A standard diagram was prepared using gallic acid (0–80 µg/mL) to estimate total phenolic content (TPC). The TPC was calculated in mg gallic acid equivalents per g dried powders of onion peel or olive leaves.

Total flavonoid content of the extracts before and after the addition of β-glucosidase was determined according to Kosalec et al. (19). Initially, 250 µL of the extract were poured into a test tube. Then, it was mixed with 750 µL of 96% ethanol, 50 µL of 10% AlCl₃ (m/V), 50 µL of 1 M sodium acetate and 1.4 mL of distilled water. After the incubation for 30 min at room temperature, the absorbance was read at 415 nm. A standard diagram was provided using quercetin standard solution (0–100 µg/mL) and total flavonoid content was expressed as mg quercetin equivalents per g dried powders of onion peel or olive leaves.

The glucose content was measured by hexokinase kit (Greiner Diagnostic GmbH, Bahlingen, Germany) before and after the enzymatic hydrolysis, which was used because the phenolic and flavonoid compounds in the onion peel and olive leaf extracts interfere with glucose in conventional methods such as the dinitrosalicylic acid (DNS) and glucose-oxidase/peroxidase assays. The effect of β-glucosidase addition to the extract was investigated as stated in the chapter above. After inactivation of the enzyme, the absorbance of the extracts with and without the enzyme treatment was read at 340 nm. The glucose concentration (mg/100 mL) was determined according to the manufacturer’s instructions.

The effect of β-glucosidase addition on the investigated extracts was determined as described in the section on total phenolic and flavonoid measurements. After the enzyme inactivation, the extracts were analyzed by loading 5 μL of the reaction mixture on TLC plates (silica gel 60 F254, 20 cm×20 cm; Merck, Darmstadt, Germany). Glucose, quercetin, quercetin-3-O-glucoside and oleuropein were used as standards. The aforementioned compounds were separated by the solvent comprising V(butanol):V(acetic acid):V(water)=3:1:1. The spots were visualized under a UV light at 254 nm (quercetin and oleuropein) or by dipping plates in a methanol solution containing 5% (V/V) sulfuric acid and heating at 120 °C for 10 min (glucose) (20). The semi-quantitative analysis of the TLC plate was done using ImageJ software (21). The TLC plate was analyzed vertically for lines related to the extracts before and after the enzyme treatment, and the results were expressed as the percentage of peak area for desired compounds (quercetin and oleuropein). Also, the TLC plate was analyzed horizontally for a semi-quantitative comparison of glucose in the extracts before and after the treatment. The results were expressed as the percentage of glucose peak area.

The effect of β-glucosidase treatment on the extracts and oleuropein standard was investigated as described in the section of total phenolic and flavonoid measurements. Oleuropein standard was applied at a concentration of 1 mg/mL. The total volume of the reaction mixture for samples and standard was 500 µL. A volume of 500 µL of ethanol was added to each extract and standard at the end of the reaction time. Oleuropein, quercetin and quercetin 3-O-glucoside standards were prepared as 50% (V/V) ethanol solutions at the concentration of 1 mg/mL. The standards and extracts were filtered and analyzed by an Agilent 1260 HPLC system (Agilent Technologies, Santa Clara, CA, USA). The separation of onion peel extract was done using the MZ PerfectSil™ 100 ODS-3 column (250 mm×4.6 mm×5 µm; MZ-Analysentechnik GmbH, Mainz, Germany). The analysis conditions were chosen according to Kim et al. (22). The mobile phase included V(water):V(formic acid)=95:5 (A) and 100% methanol (B). The binary gradient was applied as follows: 20–60% B for 0–25 min, 60–100% B for 25–25.1 min, 100–60% B for 25.1–30 min, 60–20% B for 30–30.1 min, and 20% B for 30.1–35 min. The flow rate and the injection volume were 1 mL/min and 10 µL, respectively. The absorbance was read at 360 nm. Olive leaf extract was analyzed according to Mazzei et al. (23). The mixture of V(water):V(acetonitrile)=79:21 was acidifed with o-phosphoric acid up to pH=3 and used as a mobile phase in the same column. The flow rate and the injection volume were 1.2 mL/min and 5 µL, respectively. Compounds were detected at 280 nm and ingredients in the extracts were identified by comparing their retention times with standards and literature data. Also, quercetin, quercetin-3-O-glucoside and oleuropein content in the extracts were measured before and after β-glucosidase treatment using the related standard calibration diagrams in the range 0–1 mg/mL for quercetin-3-O-glucoside and oleuropein and 0–0.5 mg/mL for quercetin. Calculations were performed based on 0.5% onion peel and 1% olive leaves extracts.

Using ANOVA, the data were analysed by SAS v. 9 software (24). The significant differences were considered using the LSD test at p<0.05. The semi-quantitative analysis of the TLC plate was carried out by ImageJ software ([REMOVED HYPERLINKF0 FIELD]21).

Mass fractions of total phenols and flavonoids were measured before and after the enzymatic hydrolysis, using gallic acid and quercetin as standards, respectively. Total phenolic mass fraction slightly increased from (16.1±0.2) to (17.5±0.7) mg/g in dry olive leaf extract and from (40.8±1.1) to (41.7±0.6) mg/g in dry onion peel extract after the enzymatic treatment. Also, flavonoid mass fraction increased after enzymatic treatment from (2.31±0.07) to (2.66±0.03) mg/g in the olive leaf extract, while it remained unchanged (16.3±0.5) mg/g in the onion peel extract. However, this increase was not significant (p<0.05). Total phenolic and flavonoid contents remain almost constant before and after the enzymatic treatment, because after the enzymatic hydrolysis, only the glucose is separated from the glycosylated compounds, producing their aglycon. Although their total amount in the medium is not changed, the content of each phenolic and flavonoid component may increase or decrease after the treatment.

Moreover, measuring the total phenolic content with the Folin-Ciocalteu method, higher values were obtained due to the unselectivity of the reagent. This reagent reacts with other reducing agents such as sugars (glucose) and phenolic compounds (32) forming blue dye with hydroxyl groups of phenols. After the enzymatic hydrolysis, blue dye intensity increases due to the separation of glucose from the glycoside compounds and an increase in free hydroxyl groups in the flavonoid ring. Cavia-Saiz et al. (29) observed a 78% increase in grapefruit juice total phenols after the treatment with naringinase. Wang et al. (33) observed a significant increase in soluble phenol and flavonoid content when treating guava leaf extract with a mixture of β-glucosidase and cellulase. In another report, the content of phenolic and flavonoid compounds in oilseed cake extract of Borago officinalis, Oenothera biennis and Nigella sativa increased after the treatment with β-glucosidase, β-glucanase, ɑ-amylase and their combinations (25). Ruviaro et al. (26) obtained similar data by treating citrus waste extract with tannase, pectinase, cellulase and β-glucosidase.

The glucose concentration increased significantly in onion peel extract from (222.2±1.2) to (247.76±0.3) mg/100 mL and in olive leaf extract from (27.0±0.7) to (75.6±0.6) mg/100 mL after the β-glucosidase treatment (2.8- and 1.11-fold for olive leaf and onion peel extracts, respectively). The high glucose concentration in olive leaf extract after enzymatic treatment is related to oleuropein hydrolysis as the main phenolic compound in olive leaves. The increase in glucose concentration in onion peel extract is associated with the hydrolysis of quercetin glucosides by β-glucosidase. Similar results were observed in the study of Ratz-Lyko et al. (25).

The results of the TLC analysis are shown in Fig. 3 and Fig. 4. According to Fig. 3a, quercetin glycosides were hydrolyzed by treating the onion peel extract with β-glucosidase. As a result, quercetin amount increased in onion peel extract after the enzymatic hydrolysis. The TLC plate analysis (Fig. 3c and 3d) confirmed the quercetin increase in onion peel extract after the enzyme action (quercetin peak area from 48.45 to 84.51%). Also, the extract treatment increased glucose amount (glucose peak area from 48.36 to 51.64%) (Fig. 3e). Choi et al. (34) obtained similar data by treating onion skin waste with cellulase, xylanase and pectinase. Also, β-glucosidase hydrolyzed oleuropein in olive leaf extract entirely (Fig. 4a), and as a result, the oleuropein was not present in the olive leaf extract after the enzymatic hydrolysis.

Moreover, glucose amount increased considerably in olive leaf extract after the enzymatic treatment (Fig. 4b). The obtained data confirmed the complete hydrolysis of oleuropein and increase of glucose (glucose peak area from 37.17 to 62.85%) in the treated olive leaf extract (Fig. 4e). Similar TLC analysis data were not found in the literature for the enzyme reaction with olive leaf extract. This is the first report of the analysis of olive leaves and onion peel extracts with a TLC plate using ImageJ software (21).

The HPLC analysis of onion peel extract showed a considerable increase in quercetin concentration and significant increase of quercetin 3-glucoside concentration after β-glucosidase treatment (Table 1). The enzyme did not hydrolyze the quercetin 3-glucoside standard (data not shown). Thus, other quercetin glucosides, including quercetin 4-glucoside and quercetin 3,4-di-glucoside were hydrolyzed by β-glucosidase. As a result, the peak obtained with the retention time of 19.65 min in the sample chromatogram before the treatment, related to quercetin 4-glucoside (9, 35), disappeared entirely after the enzymatic hydrolysis (data not shown). Similar results were observed in the mentioned literature for onion waste treated with Thermotoga neapolitana β-glucosidase. Moreover, β-glucosidase affected other glycosides in the onion peel extract.

In olive leaf extract, oleuropein (the main phenolic compound) was hydrolyzed completely by β-glucosidase within 24 h of reaction at 50 °C (Table 1). As a result, the bitterness of the extract disappeared after the enzymatic treatment. Still, the oleuropein aglycone peak did not appear in the sample chromatogram after the treatment (data not shown). The oleuropein aglycon has poor water stability. Typically, 24 h after production, it spontaneously or in the reaction with esterase converts to hydroxytyrosol and elenolic acid. Also, in the treated sample, a peak with a retention time of 5.22 min appeared, related to hydroxytyrosol (10, 36). After the treatment with β-glucosidase, the same peak was observed in the oleuropein standard chromatogram, indicating spontaneous hydrolysis of oleuropein aglycon in the water medium.

Furthermore, the enzyme affected other glycosides in the olive leaf extract. The β-glucosidase from different sources, including Streptomyces sp. (37), almond (5, 38) and olive (39, 40), has been used for oleuropein hydrolysis.

The results of different methods showed the efficacy of β-glucosidase for processing the mentioned by-products. According to the previous study (13), A. niger β-glucosidase is a stable enzyme with high storage stability, heat stability, pH stability and stability against organic solvents. Therefore, it is a potential enzyme for food waste processing, and extraction and recovery of valuable bioactive compounds with low costs. Moreover, hydrolysis of glycoside compounds into aglycone form increases their health benefits.