Analytical evaluation of phenolic compounds and minerals of Opuntia robusta J.C. Wendl. and Opuntia ficus-barbarica A. Berger

ABSTRACT In this study, 19 phenolic compounds were detected using high-throughput instrument ultra performance liquid chromatography with electrospray ionization tandem mass spectrometry (UPLC–ESI–MS/MS) in Opuntia ficus-barbarica A. Berger and Opuntia robusta J.C. Wendl. fruits. The five macro- and five micro-minerals determined in both species were analyzed using inductively coupled plasma–mass spectrometer. The phenolic compounds, mineral content, and the antioxidant capacity of the fruits of O. robusta and O. ficus-barbarica were analyzed. All phenolic compounds and minerals varied significantly between the two species. The total of phenolic compounds content was calculated as 69.237 and 66.385 mg kg−1, respectively, in O. ficus-barbarica and O. robusta. Ferulic acid was the highest quantities, 31.620 and 26.931 mg kg−1 in O. robusta and O. ficus-barbarica, in all phenolic contents, respectively. The macroelements calcium and potassium were the most abundant in both Opuntia species. The antioxidant activity of O. ficus-barbarica and O. robusta fruit samples was measured in the extracts of hexane, ethyl acetate, methanol, and water. The DPPH assay of Opuntia samples displayed a good radical scavenging inhibition, similar to butylated hydroxyanisole and butylated hydroxytoluene standards, as half maximal inhibitory concentration IC50 = 69.32 and 67.57 μg mL−1 in ethyl acetate extracts of O. ficus-barbarica and O. robusta fruits, respectively. This work presents a suitable method for the extraction, detection, and quantification of phenolic compounds by UPLC–ESI–MS/MS. MS/MS determination for multiclass determination was validated in Opuntia samples obtaining good results. Abbreviations: ABTS, 2,2′-azino-bis-3-ethylbenzthiazoline-6-sulphonic acid; AChE, acetylcholinesterase; BChE, butyrylcholinesterase; BHA, butylated hydroxyanisole; BHT, butylated hydroxytoluene; DPPH, 2,2-diphenyl-1-picrylhydrazyl; DTNB, (5,50-Dithio-bis(2-nitrobenzoic)acid; ICP/MS, inductively coupled plasma/mass spectrometer; LoD, yhe limit of detection; MRM, multiple reaction monitoring; QEs, quercetin equivalents; PEs, pyrocatechol equivalents; R2, correlation coefficients; r, Pearson’s correlation coefficient; SD, standard deviation; TIC, total ion chromatogram; UPLC–ESI–MS/MS, ultra performance liquid chromatography with electrospray ionization tandem mass spectrometry


Introduction
The cactus pear is cultivated and grown commercially in countries such as Mexico, the United States, Italy, and Israel. On the other hand, this cactus is categorized as an underutilized crop in several

Determination of individual phenolic compounds
The peel and pulp manually removed at first. Edible portions, which were separated from the seeds, were frozen at −18°C. Then, the samples were lyophilized (Christ Freeze Dryer Alpha 1-4 LD plus, Germany) in order to remove the water completely and to reduce to a fine dried powder. The lyophilized powder (3 g) was extracted by 30 mL of 80% (v/v) acetone at 25°C for 6 h, and then, ultrasonic extraction was applied for 15 min. The extract was centrifuged at 4000 rpm for 10 min and then filtered using Whatman No. 4. After that, the residue was extracted with three additional 30 mL portions of 80% (v/v) acetone, then the extracts were combined, and the solvent in the extracts was evaporated at 40°C. That extract was treated with 30 mL of hexane for tree times in order to remove fatty acid content and that of hexane phase is discarded. The solid residue was redissolved in methanol and filtered through a PTFE-20/25 LC filter disk as reported by a previous method with slight modification. [29] Phenolic compounds of Opuntia samples were analyzed using high-throughput instrument a Waters ultra performance liquid chromatography with electrospray ionization tandem mass spectrometry (UPLC-ESI-MS/MS) and C18 column (Acquity UPLC BEH C18 100 mm × 2.1 mm, 1.7-μm particle size) and the separation of compounds was performed by gradient elution. The mobile phases were composed of solvent A (0.1% acetic acid in water) and solvent B (0.1% acetic acid in acetonitrile), and the flow rate was 0.650 mL/min, and at 40°C column temperature. The liquid chromatography and mass spectrometry conditions of the analysis [30] were displayed in Table 1.

Determination of metal contents using ICP-MS
Opuntia fruit samples were washed with ultrapure water in order to clean residual portions of soil, and then samples were dried. Samples (0.2 g) were weighted into teflon tube and mixed with nitric acid (3 mL), hydrochloric acid (0.5 mL), and hydrogen peroxide (0.5 mL) and then extracted with microwave method (Cem Mars 5, Matthews, NC, USA) at a 1200-W power. After the extract solution was cooled down, it was added to flask (20 mL) by completing volume with ultrapure water. Opuntia fruit samples were analyzed for the assessment of metal and mineral content using an Agilent inductively coupled plasma-mass spectrometer (ICP-MS) (Agilent 7700×, Tokyo, Japan). Calibration ranges for the analysis were 0.5-200 ppb for microelements and 0.5-25 ppm for macroelements. Operation parameters for the analysis were as follows: plasma power, 1550 W; plasma mode, normal; plasma gas flow rate, 15.0 L min −1 ; auxiliary gas flow rate, 1.0 L min −1 ; carrier gas flow rate, 0.89 L min −1 ; dilution gas flow rate, 0.15 L min −1 ; sample depth, 8.0 mm; spray chamber temperature, 2°C; kinetic energy discrimination, 3 V; helium gas flow rate, 4.5 mL min −1 .

Measurement of in-vitro antioxidant capacity
Inhibition of β-carotene bleaching assay The total antioxidant activity was determined using β-carotene-linoleic acid test system based on the detection of inhibition of conjugated dien hydroperoxides due to oxidation of linoleic acid. [31] β-Carotene (0.5 mg), dissolved in 1 mL of chloroform, was mixed with linoleic acid (25 μL) and Tween 40 emulsifier (200 mg). Chloroform was evaporated under vacuum; 50 mL of distilled water saturated with oxygen was added by vigorous shaking. Aliquots (160 μL) of this emulsion were added to 40 μL of the extract solutions at different concentrations. As soon as the emulsion was added to each tube, the zero time absorbance was initially measured at 470 nm using 96-well microplate reader Spectra Max 340PC Molecular Device, and then, the absorbance measurements were done for every 30 min until 120 min. The results were given as 50% inhibition concentration (IC 50 ). The sample concentration inhibiting 50% antioxidant activity (IC 50 ) was calculated from the graph of activity percentage against sample concentration.

DPPH radical scavenging activity
The free radical scavenging activity of Opuntia extract was determined using DPPH free radical according to literature. [29] The extract solutions with different concentrations (40 μL) and ethanolic solution (120 μL) containing DPPH radicals (0.4 mM) were incubated at room temperature in darkness for 30 min. Absorbance was measured at 517 nm using 96-well microplate reader Spectra Max 340PC Molecular Device.
ABTS cation radical decolorization assay ABTS cation radical decolorization assay was analyzed according to Thaipong et al. [32] The ABTS (7 mM) in water and potassium persulfate (2.45 mM) reacted to give ABTS •+ , stored in the dark at room temperature for 12 h, and oxidation of ABTS appeared immediately; however, the stability of absorbance was gained after 6 h. Then, the sample solution (40 μL) in ethanol at different concentrations was mixed with ABTS •+ solution (160 μL), giving the absorbance at 734 nm using 96-well microplate reader Spectra Max 340PC Molecular Device. Table 1. Accuracy, precision, linearity, sensitivity of evaluation of phenolic compounds, and chromatography-mass spectrometry conditions [30] .

Statistical analysis
The results were reported as mean and standard deviation of the mean. Data were subjected to analysis of variance (ANOVA). The least significant difference test was applied to mean to the mean values using the STATISTICA for Windows release 5.0, and correlation coefficients (r) were computed to establish relationships between phenolic compounds and minerals of fruits.

Result and discussion
Accuracy, precision, linearity, sensitivity of evaluation of phenolic compounds Linearity was determined for phenolic compounds in the concentration range of 0-100 mg kg −1 .
Correlation coefficients (R 2 ) of linear regression analysis from calibration curves displayed >0.93, between 0.937176 and 0.999939 (Table 1). The limit of detection for sensitivity was evaluated in the range of 0.005-0.020 mg kg −1 and accuracy was determined by evaluating recovery values for each phenolic compound. The chromatography and mass spectrometry parameters [27] values were shown in Table 1.

Phenolic compounds determination
The individual phenolic compound values for each cactus pear are presented in Table 2. One-way ANOVA of the data showed that the effects of cactus pear species were statistically significant Twelve (p-hydroxybenzoic acid, vanillin, gentisic acid, protocatechuic acid, gallic acid, ferulic acid, homogentisic acid, myricetin, pyrogallol, rutin, pyrocatechol, and 3,4-dihydroxy benzaldehyde) among the detected phenolic compounds contents were found in O. robusta higher than those of O. ficus-barbarica. However, p-coumaric acid, vanillic acid, chrysin, caffeic acid, luteolin, naringenin, and quercetin were detected in O. ficus-barbarica higher amount than O. robusta. As the other plant species [9] have high phenolic content, the studied Opuntia species can be considered not only as rich sources of phenolics and flavonoids but as promising sources of natural antioxidants as well. The total of phenolic compounds amount was calculated as 69.237 and 66.385 mg kg −1 , respectively, in O. ficus-barbarica and O. robusta.
The results show that 3,4-dihydroxy benzaldehyde contents were the lowest phenolic compound which was found 0.063 ± 0.005 mg kg −1 in O. ficus-barbarica and 0.098 ± 0.007 mg kg −1 in O. robusta. Additionally, quercetin, chrysin, naringenin in O. robusta and pyrocatechol, naringenin, chrysin, rutin, vanillin in O. ficus-barbarica were detected in low amounts. As it is presented in Table 2, ferulic acid was the highest quantities, 31.620 ± 0.018 and 26.931 ± 0.022 mg kg −1 in O. robusta and O. ficus-barbarica, in all phenolic contents, respectively. In both, Opuntia species have quite similar in the total of phenolic content; however, p-coumaric acid displayed exceptional results that the detected content in O. ficus-barbarica was sixfold higher than O. robusta.

Mineral content determination
The minerals determined in both species were analyzed using ICP-MS and the results were displayed in Table 3. For each Opuntia species, five macroelements (Na, Ca, K, P, Mg) and five microelements (Fe, Mn, Zn, Mo, Cu) were determined. Generally, the macroelements calcium and potassium were the most abundant in both Opuntia species. The effects of Opuntia species were statistically significant (p < 0.05) in the content of all minerals. The total mineral content was determined 907.829 mg 100 g −1 for O. ficus-barbarica and 1067.882 mg 100 g −1 for O. robusta. Also, total macroelements content was occurred 865.125 mg 100 g −1 for O. ficus-barbarica and 1044.035 mg 100 g −1 for O. robusta. On the other hand, O. robusta contained less total microelement content than that of O. ficus-barbarica. The total microelement content was 42.183 and 23.084 mg 100 g −1 for O. ficus-barbarica and O. robusta, respectively. The most abundant macroelement in Opuntia species was calcium, which was detected 639.128 ± 1.457 mg 100 g −1 in O. robusta and 510.941 ± 4.702 mg 100 g −1 in O. ficus-barbarica. Potassium and magnesium come in the second and third order in both Opuntia species. The content of potassium and magnesium was 240.347 ± 3.674 and 89.420 ± 1.155 mg 100 g −1 in O. ficusbarbarica, on the other hand, 281.869 ± 2.905 and 108.804 ± 0.359 mg 100 g −1 in O. robusta. Phosphorus was the mineral present in the lowest concentration, which was 0.110 ± 0.004 mg 100 g −1 in O. ficus-barbarica and 0.138 ± 0.005 mg 100 g −1 in O. robusta.
Nutritional quality of daily food supply, especially with respect to essential nutrient minerals, such as magnesium, iron, and zinc, could be an important goal of leafy vegetable crops. [34] These essential nutrient minerals can be fulfilled with the studied Opuntia species.

Correlation coefficient between parameters
Pearson's correlation coefficient (r) was obtained from correlation analysis and used to describe the correlations among 19 phenolic compounds in both O. robusta and O. ficus-barbarica. The correlation coefficients were determined among phenolic compounds for O. robusta. The highest correlation coefficient was significantly found between p-hydroxy benzoic acid and chrysin as 0.959. The other high correlation coefficients were calculated between vanillin and caffeic acid (0.868), naringenin and quercetin (0.867), caffeic acid and ferulic acid (0.858), and gentisic acid and chrysin (−0.843). p-Hydroxy benzoic acid was significantly and positively correlated with vanillic acid, chrysin, myricetin, and pyrocatechol and negatively correlated with gentisic acid.
Pyrocatechol content of O. robusta fruit was significantly and positively correlated with six of other phenolic compounds: p-hydroxy benzoic acid, gentisic acid, chrysin, naringenin, pyrogallol, quercetin, and 3,4-dihydroxy benzaldehyde. Myricetin was significantly correlated with vanillic acid (0.827), chrysin (0.785), gentisic acid (0.673), and p-hydroxybenzoic acid (0.663). Additionally, rutin was also was correlated with four of detected phenolics: ferulic acid, luteolin, naringenin, and caffeic acid. The coefficients of correlation were evaluated among minerals in O. robusta fruit. Zinc content of O. robusta was significantly correlated with all minerals except molybdenum. The highest correlation coefficient was observed between calcium and potassium (−0.921). One of highest correlation coefficients was found also with zinc and manganese (0.911).
Calcium, phosphorus, iron, manganese, magnesium, zinc, and copper were significantly correlated with each other. Among these minerals, the highest coefficient of correlation was observed between iron and magnesium (−0.894). Sodium was significantly correlated with phosphorus, zinc, and copper. Molybdenum was only correlated with potassium and magnesium. Simple correlation coefficients between phenolic compounds and each mineral detected in O. robusta samples. There are no statistically significant (p = 0.05) relationship among p-coumaric acid, pyrogallol, quercetin, pyrocatechol, and 3,4-dihydroxy benzaldehyde with all determined mineral contents.
Luteolin showed significant positive correlations between calcium and magnesium and negative correlations between potassium, phosphorous, iron, manganese, and zinc. The highest correlation coefficient was observed between luteolin and manganese (−0.891). p-Hydroxy benzoic acid and naringenin were shown significant correlation with only iron. The other high and significant correlation coefficients were found between vanillic acid and magnesium and iron (−0888 and 0840), between luteolin and zinc (−0.880), and between myricetin and magnesium as −0.836. Gentisic acid, protocatechiuc acid, chrysin, gallic acid, and rutin were significantly correlated with a few minerals such as manganese, magnesium, sodium. Overall, mostly moderate and weak correlation coefficients were observed between phenolic compounds and minerals in O. robusta.
Phenolic compounds of O. ficus-barbarica determined in this study were individually correlated with each other. In general, more significant correlation coefficients were found among phenolic compounds in O. ficus-barbarica than observed in O. robusta. The highest significant correlation coefficient was found between vanillin and myricetin (r = 0.976, p = 0.05). Second and third high correlation coefficients were observed between vanillin and pyrocatechol, and between myricetin and pyrocatechol as −0.963 and −0.926, respectively. Also, high and significant correlation coefficient was calculated between vanillic acid and quercetin (0.902). Vanillin also was negatively correlated with gallic acid and positively correlated with luteolin.
The other high correlation coefficients (above 0.800) were found between myricetin and luteolin and gallic acid, between naringenin and 3,4 dihydroxy benzaldehyde, between pyrogallol and gallic acid, and between gentisic acid and protocateghuic acid. Rutin, naringenin, feulic acid, and gallic acid were significantly correlated with only one phenolic compound detected in this study.
The correlation coefficients among minerals in O. ficus-barbarica fruits are evaluated. Potassium, phosphorus, iron, magnesium, zinc, and molybdenum were significantly correlated with all minerals except copper, which was correlated only with calcium. The highest correlation coefficient was observed between zinc and molybdenum (−0.902). Second and third high correlation coefficients were found between magnesium and phosphorus (0.892), and between potassium and phosphorus (−0.889) in detected minerals for O. ficus-barbarica. Also iron with phosphorus (0.876), and magnesium with manganese (−0.874) and potassium (−0.872), showed high correlation coefficients.
Overall, strong correlations were observed among all minerals in O. ficus-barbarica fruit. The results of statistical analysis of all phenolic compounds and minerals of O. ficus-barbarica fruit are evaluated. Correlation coefficients (r) exhibited significant differences at the 0.05 probability level, ranging from 0.601 to 0.853. The highest r value was found between gallic acid and potassium. Gallic acid showed the positively and negatively significant correlation coefficients with all minerals except sodium and copper. Also, pyrogallol showed high correlations with phosphorus (−0.840) and manganese (0.838).
There is no statistically significant relationship among p-coumaric acid, vanillic acid, ferulic acid, naringenin, and pyrocatechol with all detected mineral in O. ficus-barbarica fruits. In addition, p-hydroxy benzoic acid, vanillin, gentisic acid, chryisin, caffeic acid, luteolin, myricetin, and 3,4-dihydroxy benzaldehyde had no significant relationship with minerals except a few. In general, moderate correlations were observed between phenolic compounds and detected minerals in O. ficus-barbarica fruits.

Antioxidant activity of O. ficus-barbarica and O. robusta fruits
The antioxidant activity of O. ficus-barbarica and O. robusta fruit samples was measured in the extracts of hexane, ethyl acetate, methanol, and water. In terms of lipid peroxidation inhibition, the β-carotenelinoleic acid assay of hexane extracts of O. ficus-barbarica and O. robusta fruit samples (IC 50 = 9.56 ± 0.48 and 8.03 ± 0.22 μg mL −1 , respectively) revealed activity close to standards. The results were presented in Table 4. In the ABTS cation radical scavenging activity, ethyl acetate extracts of O. ficus-barbarica and O. robusta fruit samples showed moderate activity as SC 50 = 23.47 ± 0.53 and 21.75 ± 1.01 μg mL, respectively, but still lower than standards. Both of the assays displayed less scavenging inhibition compared to standards in the extracts of hexane, methanol, and water.
In the case of DPPH radical scavenging activity, the DPPH assay of Opuntia samples displayed a good radical scavenging inhibition, similar to BHA and BHT standards, as half maximal inhibitory concentration SC 50 = 69.32 ± 0.96 and 67.57 ± 0.83 μg mL in ethyl acetate extracts of O. ficus-barbarica and O. robusta fruits, respectively. The extracts of ethyl acetate, methanol, and water exhibited limited activity, compared to BHA, BHT, and α-tocopherol standards. Thus, results of antioxidant activity of analysis of studied Opuntia fruits performed a good correlation between antioxidant activity and phenolic content. All activity data were displayed in Table 4.

Anticholinesterase activity of O. ficus-barbarica and O. robusta fruits
The BChE activity of the studied O. ficus-barbarica and O. robusta fruits of ethyl acetate extract exhibited higher activity, compared with galanthamine, as IC 50 = 51.05 ± 1.25 and 46.27 ± 1.57 μg mL, respectively. On the other hand, all four extracts of O. ficus-barbarica and O. robusta fruit samples showed almost no activity against AChE (Table 4). To the best of our knowledge, there are some studies about total phenolic contents and minerals of flowers, cladodes, seeds, and juices of O. ficus-indica and a few other Opuntia spp. [11,12,14] Despite an extensive literature search, information on the individual phenolic compounds and mineral contents of O. robusta and O. ficus-barbarica, which were detected using by UPLC-ESI-MS/MS and ICP/MS, respectively, are not available in the literature.

Conclusion
In this study, it is aimed to investigate the individual phenolic compounds, mineral content, and the antioxidant capacity of O. robusta and O. ficus-barbarica fruit samples. After the evaluation of the results, the studied Opuntia species displayed rich sources of phenolic profile, mineral content, and also promising sources natural antioxidant. Phenolic profile composition and mineral content varied among the species. Therefore, the chemical composition was used successfully to obtain information about the relationship among Opuntia species. It seems that O. robusta and O. ficus-barbarica fruits can supply a dietary intake of essential nutrients. This research shows the potential of fruits of Opuntia species. Those can be used as pulp and juices, which have an important source of natural antioxidants and nutraceuticals.