High resolution LC-MS characterization of phenolic compounds and the evaluation of antioxidant properties of a tropical purple radish genotype

2.3.1 Plant sample preparation

The root samples were divided into the periderm and xylem, and the leaf samples were divided into petiole and lamina. For extraction, each sample was thoroughly homogenized under fresh condition. The homogenized sample (5 g) was drawn in a 50 mL amber-colored polypropylene centrifuge tube with a quality control standard (Forchlorfenuron, 1 mg/L), and immediately extracted with methanol (20 mL, 1% formic acid) by vortexing at 2000 rpm for 2 min in the dark, followed by centrifugation at 10000 rpm for 10 min at 10 °C. The supernatant was diluted with water (1:2, v/v), filtered through 0.2 μm PTFE membrane, and finally injected (5 μL) into the LC-MS.

2.3.2 LC-MS [UPLC-(ESI)-QToFMS] conditions

The analysis was performed on an Acquity Ultra Performance Liquid Chromatograph (UPLC), coupled to a QToF-MS (Synapt G2 HDMS, Waters Corporation, Manchester, UK). The QToF-MS was operated with electrospray ionization (ESI) at the mass resolution of 20000, and controlled by MassLynx 4.1 software. which simultaneously provided quick switching from low energy scan at 4 V (full scan MS) to high energy scan (10–60 V ramping) during a single LC run. The low-CE experiments provided information about the intact molecular ion (e.g. M⁺, [M+H]⁺), while the high-CE scan generates fragment ion information. The source parameters were set as follows: capillary 3 kV, sampling cone 30 V, extraction cone 5 V, source temperature 120 °C, desolvation temperature 500 °C, desolvation gas flow 1000 L/h, and cone gas flow 50 L/h. The mass spectrometer was calibrated with 0.5 mM of sodium formate. The lock spray and the reference mass leucine enkephalin (m/z 556.2771 in positive and 554.2670 in negative polarity) was used for the mass correction with a flow rate of 10 μL/minute at the concentration of 2 μg/mL at an interval of 20 s. At 35 °C, the chromatographic separation was performed on an Acquity UPLC BEH C18 column (2.1 × 100 mm, 1.8 μm, Waters India Pvt. Ltd., Bengaluru). The mobile phase consists of A: methanol:water (10:90, v/v) and B: methanol:water (90:10, v/v) with 0.1% formic acid in both the phases. A gradient program was used with 0.4 mL/min flow rate, with 0–0.5 min/90% A, 4.5 min/50% A, 4.5–8 min/50–2%A, 8–11 min/2% A, 11–11.5 min/2–90% A, 12–15 min/90% A. The total ion chromatograms (TICs) for the acquired samples i.e. leaf lamina, leaf petiole, root xylem, and root periderm are given in the Supplementary Fig. S1.

2.3.3 Data analysis

The LC-MS data files (n = 6) were processed by the UNIFI software (version 1.7, Waters Corporation) with a screening solution workflow which helped in automated data processing to reporting of the positive identifications by comparison with a database of polyphenols and their derivatives. This database was developed for more than 1200 compounds by collecting the compound specific information (chemical structure, molecular formula, molecular mass) from various web-based resources (e.g. Chemspider) and published research papers. Based on the database information, the compounds were identified with mass errors of <5 ppm for the precursor as well as one or more product ion. The UNIFI software has a unique database feature which allows the users to quickly add any new compounds and create a customized library of compounds that includes detailed structural information (Deng et al., 2016). These product ions generated through collision induced dissociation were matched against the theoretical fragmentation pattern. The identified compounds were relatively quantified against quercetin and pelargonidin-3-O-glucoside, which were used as the external reference standards at 2 µg/mL concentration for the non-anthocyanins and anthocyanin derivatives, respectively. From these relative concentrations, the percentage distribution of various classes of phenolic compounds in the root and leaf fractions was estimated.

2.4.1 Sample preparation

For the evaluation of antioxidant activity, root periderm, root xylem, leaf petiole and leaf lamina were separately homogenized into pulp, and extracted with ethanol (80%) by stirring for 30 min, and centrifuged for 15 min at 13,000 rpm at 4 °C. The supernatant was collected in a test tube, and the residue was re-extracted twice with 80% ethanol, and then centrifuged. All the supernatants were then collected and pooled together. Before further analysis, the extracts were preserved at −20 °C.

Four in vitro methods were used for the analysis of antioxidant activities. The metal reducing potential of radish extracts was evaluated using the ferric reducing antioxidant power (FRAP) and cupric reducing antioxidant capacity (CUPRAC) methods, while the free radical scavenging assay was performed using 2,2-diphenyl-1-picrylhydrazyl (DPPH), and trolox equivalent antioxidant capacity (TEAC) assays.

2.4.2 FRAP

For the FRAP method (Benzie and Strain, 1996), the reagent was prepared by mixing 300 mM acetate buffer (pH 3.6), 10 mM TPTZ [2,4,6-Tripyridyl-s-triazine] in 40 mmol HCl and 20 mmol FeCl₃ in the ratio 10:1:1 (v/v/v). An amount of 3 mL of the FRAP reagent was mixed with 100 μL of aliquot in a test tube and vortexed in an incubator at 37 °C for 30 min in a water bath. Reduction of ferric tripyridyltriazine to the ferrous complex formed an intense blue color which was measured through a UV–Visible spectrophotometer (Shimadzu UV 1601) at 593 nm at the end of the 4 min. Results were expressed in terms of μmol Trolox Equivalent (TE)/g fresh weight (fw).

2.4.3 CUPRAC

In this method (Apak et al., 2008), an aliquot (100 μL) was mixed with 1 mL each of CuCl₂ (1.0 × 10⁻² mol L⁻¹), neocuproine alcoholic solution (7.5 × 10⁻³ mol L⁻¹), ammonium acetate (1 mol L⁻¹, pH 7.0) buffer solution and 1 mL of water to make the final volume to 4.1 mL. After 30 min, the absorbance was recorded at 450 nm against the reagent blank. A standard curve was prepared using different concentrations of Trolox (100–2000 μM). The results were expressed as μmol TE/g fw.

2.4.4 DPPH

The DPPH assay is based on the measurement of the scavenging ability of antioxidants towards the stable DPPH radical (Brand-Williams et al., 1995). An aliquot (100 μL) of the radish extract was added to 3.9 mL DPPH in methanol (95%) and shaken vigorously. Any change in the absorbance of the sample extract was measured at 515 nm for 30 min until the absorbance reached a steady state. The percentage inhibition of DPPH was calculated by the following formula: $$\%\text{Inhibition}=\frac{A0-A}{A0}\times 100$$ where A0 is the initial absorbance, and A is the final absorbance of the sample extract measured at 515 nm. Methanol (95%) was used as the blank. The calibration curve between% inhibition and concentration of Trolox (100–2000 μM) was then established. Results were expressed as μmol TE/g fw.

2.4.5 TEAC

The TEAC method that was proposed by Re et al. (1999) was used. The ABTS^•+ free radical (7 mmol) was prepared by mixing 7 mmol ABTS with 2.45 mmol of potassium persulphate, which acted as an oxidizing agent. An aliquot (10 μL) of the extract was added to 90 μL of ABTS^•+ solution and the absorbance was measured at 734 nm (at 30 °C), exactly 10 min after initial mixing. The percentage inhibition of ABTS^•+ was calculated in the same way as for the DPPH method. The radical-scavenging activity of the test samples was expressed as trolox equivalent antioxidant capacity (TEAC μmol TE/g fw).

3.1.1 Flavonol

The putatively identified flavonols included 12 kaempferol glycosides, 6 quercetin glycosides, 2 isorhamnetin glycosides, 1 spinacetin glycoside, and 1 methyl galangin. The identifications were based on certain criteria. For example, kaempherol-3-O-rhamnoside (Table 1, Sr. 1) was identified based on the protonated molecular ion m/z 433.1137 (mass error 2.11 ppm) with elemental composition C₂₁H₂₀O₁₀ and two characteristic fragment ions (m/z 153.0177 and m/z 287.0548) (Fig. 1a). Kaempherol-3-O-rhamnoside (Table 1, Sr. 1 and 2) appeared at two different retention times with baseline separation which indicated that this compound probably appeared in two isomeric forms. Quercetin-3-O-rhamnosyl-galactoside (Table 1 sr. 15) was identified based on the protonated molecular ion m/z 611.1592 (mass error -1.72 ppm), elemental composition C₂₇H₃₀O₁₆ and characteristic fragment ions with m/z 449.10781 and 303.04929 (Fig. 1b). In a similar manner, the other compounds (Table 1) were identified. Among the flavonols, 11 were non-acylated glycosides and 12 were acylated glycosides. The acylated flavonols could be identified based on the increase in the mass of the corresponding precursor ions. They were either acylated singly (10 compounds), or triply (2 compounds) with hydroxycinnamic acid (p-coumaric acid, caffeic acid, ferulic acid, or sinapic acid) at the 3-position of the C ring or aliphatic acid (malonic acid) at the 7-position of the A-ring. The non-acylated flavonols were mostly identified in the leaf fraction; however, all the acylated flavonols were detected in the leaf, as well as in the root fractions.

Close

Fig. 1 Product ion mass spectra for (1a) Kaempferol-3-O-rhamnoside, (1b) Quercetin-3-O-rhamnosyl-galactoside and (1c) Cyanidin-3-O-(p-coumaroyl-feruloyl)-sophoroside-5-(malonyl) glucoside.

Export to PPT

Except kaempferol-3-O-p-coumaryl-shophoroside-7-O-glucoside (compound 8, Table 1) none of the other flavonols was earlier reported either in radish root or leaf (Kamil and Kalina, 1977; Papetti et al., 2014). Among the identified flavonols, 7 similar compounds were earlier reported in mustard, broccoli and Chinese cabbage (Lin and Harnly, 2010), and 13 similar flavonols were also recorded in red mustard greens (Lin et al., 2011b).

3.1.2 Anthocyanins

Similar to flavonols, the anthocyanins were putatively identified based on their precursor and one or more product ions, each with <5 ppm of mass errors. For example, Cyanidin-3-O-(p-coumaroyl-feruloyl)-sophoroside-5-(malonyl) glucoside was identified based on the observed accurate mass (m/z 1181.2986) with 2.35 ppm of mass error and its identity was further confirmed by the product ions with m/z of 287.0547, 535.1087, and 1151.29009 (Fig. 1c). The twenty-eight putatively identified anthocyanins included 22 cyanidin glycosides, 5 pelargonidin glycosides, and 1 delphinidin glycoside. Among these, 6 were non-acylated and 22 were acylated glycosides. The non-acylated anthocyanins were mostly identified in the leaf fraction, whereas all the acylated anthocyanins were detected both in the leaf and root fractions. The predominant structural backbones of the acylated anthocyanins were either anthocyanin-3-O-diglucoside-5-O-glucoside, or anthocyanin-3-O-triglucoside-5-O-glucoside, or anthocyanin-3-O-diglucoside-5-O-diglucoside. These are specific to the family Brassicaceae (Lin et al., 2011a). In addition, 2 mono-acylated galactosides were also putatively identified (compounds 29, 30, Table 1).

Hydroxycinnamic acids (p-coumaric/caffeic/ferrulic/sinapic acid) were attached with the C3 of the flavynium ring and aliphatic acids (either malonic or succinic acid) were attached to the C5 of the flavylium ring. The acylated anthocyanins included 6 mono-acylated, 12 di-acylated, and 9 tri-acylated compounds, some of which were also reported earlier in the root of purple radish varieties: ‘Benikanmi’ (Tatsuzawa et al., 2010), ‘Bordeaux’ (Lin et al., 2011a), and the radish sprout (Baenas et al., 2015). Similarly, some of these compounds were also earlier reported in the leafy vegetables belonging to the Brassicaceae family, e.g., red mustard greens (Lin et al., 2011b).

Anthocyanins that are reported for the first time includes the compounds 29, 30, 31, 34, 35, 36, 38, and 41 (Table 1). In this experiment, along with cyanidins, a few pelargonidins and one delphinidin were also identified. The findings are in agreement with Baenas et al. (2015), who reported the presence of pelargonidin, delphinidin and peonidin along with cyanidin in radish sprout. The presence of these compounds indicates possible mixing of the genes from pink and purple radish. Since radish is highly cross-pollinated in nature with a complex nature of pigment inheritance, the possibilities of occurrence of all of these pigments in the purple genotype are justifiable.

3.1.3 Other compounds

In addition to flavonols and anthocyanins, a dihydroflavonol, 4 flavones, 2 isoflavonoids, 3 phenolic acid, and a hydroxybenzaldehyde were putatively identified based on matching with the compound database through UNIFI. Except 6-prenylnaringenin and dihydro-caffeoyl-3-O-glucouronide, all the other compounds were detected only in the leaf samples. Among these compounds, except coumaric acid, which was earlier reported by Goyeneche et al. (2015), none has any mention in the literature. However, the occurence of dihydroflavonol (Marles et al., 2003), and 1,2-dihdroxy-feruloyl-gentibiose were reported in other crops of the Brassicaceae family (Lin et al., 2011b). Many of the compounds reported here are known for their potential health benefits, including antioxidant, anti-diabetic, anti-hypertensive, anti-inflammatory, and anti-cancer activities. (Tapas et al., 2008).