3205
J. Sep. Sci. 2014, 37, 3205–3213
Mohammad B. Hossain1 Gabriel Camphuis2 1 ´ Ingrid Aguilo-Aguayo Nirupama Gangopadhyay1 Dilip K. Rai1 1 Department
of Food Biosciences, Teagasc Food Research Centre, Ashtown, Dublin, Ireland 2 Polytechnique Institute of LaSalle, Beauvais, France Received May 30, 2014 Revised July 29, 2014 Accepted August 11, 2014
Research Article
Antioxidant activity guided separation of major polyphenols of marjoram (Origanum majorana L.) using flash chromatography and their identification by liquid chromatography coupled with electrospray ionization tandem mass spectrometry† Marjoram extracts have been separated into polar and nonpolar parts using liquid–liquid extraction. Both polar and nonpolar parts of the extracts were further fractionated by flash chromatography. The obtained fractions (90 polar and 45 nonpolar fractions) were investigated for their antioxidant activities by 2,2-diphenylpicrylhydrazyl and ferric ion reducing antioxidant power assays. A direct, positive, and linear relationship between antioxidant activity and total phenolic content of the fractions was observed. Based on antioxidant and total phenolic content data, the three fractions with the high antioxidant activities from polar and nonpolar part of the extract were analyzed for their constituent polyphenols by liquid chromatography coupled with electrospray ionization tandem mass spectrometry. Compounds were identified by matching the mass spectral data and retention time with those of authentic standards. Identification of the compounds for which there were no “in-house” standards available was carried out by accurate mass measurement of the precursor ions and product ions generated from collision-induced dissociation. Rosmarinic acid was found to be the strongest antioxidant polyphenol conferring the highest antioxidant activity to fractions 47 and 17 of polar and nonpolar part of the extract, respectively. The identification of the rosmarinic acid was further confirmed by 1 H NMR spectroscopy. Keywords: Antioxidant activity / Flash chromatography / Liquid–liquid partitioning / Polyphenols / Tandem mass spectrometry DOI 10.1002/jssc.201400597
1 Introduction In the past decades, scientific research interest has increased considerably in naturally occurring antioxidants for foods or medicinal applications since the use of synthetic antioxidants such as butylated hydroxyanisole and butylated hydroxytoluene is being restricted due to health and safety concerns [1, 2]. Natural antioxidants can protect the human body from free radicals and could retard the progress of many chronic diseases as well as lipid oxidative rancidity in foods [3–5]. A host of potentially beneficial physiological effects have Correspondence: Dr. Mohammad B. Hossain, Department of Food Biosciences, Teagasc Food Research Centre, Ashtown, Dublin 15, Ireland E-mail: [email protected] Fax: 00353-0-18059550
Abbreviations: CID, collision-induced dissociation; DPPH, 2,2-diphenylpicrylhydrazyl; FCR, Folin–Ciocalteu reagent; FRAP, ferric ion reducing antioxidant power; Trolox, (6hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid); TSP, 3-(trimethylsilyl)propoinic acid-d4
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
been postulated for natural antioxidants over the past three decades, which are supported by extensive animal studies. Among these are beneficial influences on lipid metabolism, efficacy as antidiabetic, ability to stimulate digestion, antioxidant property, anticarcinogenic, and anti-inflammatory properties [6, 7]. Oxidation of polyunsaturated fatty acids not only lowers the nutritional value of food [8], but is also associated with cell membrane damage, aging, heart disease, and cancer in living organisms [9]. Therefore, the addition of natural antioxidants to food products has become popular as means of increasing shelf life and to reduce wastage and nutritional losses by inhibiting and delaying oxidation [10]. Several studies reported that extracts of Origanum majorana (marjoram) had high antioxidant capacity [11–13] mostly due to the polyphenolic compounds present in them. A total of 31 polyphenols
†This paper is included in the virtual special issue sample preparation in mass spectrometry available at the Journal of Separation Science website. Colour Online: See the article online to view Fig. 3 in colour.
www.jss-journal.com
3206
M. B. Hossain et al.
were identified in marjoram in a previous study [14]. However, to the best of our knowledge, no study has demonstrated which of these polyphenols of marjoram were mainly responsible for its high antioxidant activity. Therefore, an antioxidant activity guided fractionation approach is followed to establish this relationship in the present study. A rapid and easyto-use solvent-based partitioning in combination with automated flash chromatographic fractionation has been used to separate the most antioxidant fractions. The importance of these sample preparation techniques and various other sample cleanup methods for analyses of polyphenols has been extensively reviewed by Stalikas [15]. Subsequent structural characterization and quantification of major polyphenols in Origanum majorana by LC–ESI-MS/MS have been presented.
2 Materials and methods 2.1 Samples and reagents Dried and ground marjoram was provided by AllinAll Ingredients, Dublin, Ireland. According to product specifications, the country of origin of the spice used was Turkey. The spice was air-dried after heat treatment (steam sterilization at 120⬚C for 30 s). The dried spice was ground (particle size range 500–600 m) and stored at −20⬚C in darkness. Three polyphenol standards namely epigallocatechin, quercetin, and rosmarinic acid, and an NMR internal reference 3-(trimethylsilyl)propoinic acid-d4 (TSP) sodium salt were purchased from Sigma–Aldrich, Wicklow, Ireland. Two flavonoid standards, apigenin and luteolin-7O-glucoside, were purchased from Extrasynthese, France. HPLC-grade methanol, acetonitrile, ethylacetate, and water were purchased from VWR International, Leicestershire, UK and Lennox Laboratory Supplies, Dublin, Ireland, respectively. The purity of standards and solvents were in the range of 95–99.8%. Folin–Ciocalteu reagent (FCR), gallic acid, sodium carbonate (Na2 CO3 ), hydrochloric acid (HCl), formic acid, deuterated phosphate buffer, 2,2-diphenylpicrylhydrazyl (DPPH), sodium acetate anhydrous, ferric chloride hexahydrate, 2,4,6-tri(2-pyridyl)-s-triazine, and 6-hydroxy-2,5,7,8tetramethylchroman-2-carboxylic acid (Trolox) were purchased from Sigma–Aldrich. 2.2 Preparation of solid/liquid extract and its fractionation Antioxidant-guided fractionation of marjoram extracts was carried out as shown in Fig. 1. Dried and ground spice samples (50 g) were sequentially extracted twice using 500 mL of 80% aqueous methanol, each time in the dark at room temperature (23⬚C). The sample suspension was shaken overnight in an orbital shaker (MaxQ 6000 Shaker, Thermo Fisher Scientific, MA, USA) set at 150 rpm at room temperature. The mixture was then filtered through a Buchner funnel (1 m). The extracts were dried immediately under reduced pressure using a rotary evaporator (Heidolph C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Sep. Sci. 2014, 37, 3205–3213
Rotary Evaporator, Germany) with a water bath set at 50⬚C. This dried extract was redissolved in water (500 mL) and partitioned against ethylacetate (500 mL). The aqueous polar part contained 89.6% w/w of the methanolic extract. This extract was freeze-dried and resuspended in a minimal amount of water (60 mL), which was further fractionated using RP flash chromatography (Varian IntelliFlash 310, CA, USA). Flash chromatography was performed on a SuperFlashTM C18 column with a sorbent mass of 300 g and average size of the particles was 40–60 m. A stepwise gradient from 10% aqueous methanol to 90% methanol in 45 min at a flow rate of 40 mL/min was used to separate the polyphenols of the polar extracts (Fig. 2). A total of 90 fractions were generated following a time scale of 0.5 min/fraction. Meanwhile the nonpolar ethylacetate extract was dried under reduced pressure as mentioned for methanolic extract. The dried nonpolar extract was also redissolved in minimal amount of fresh ethyl acetate (50 mL) and chromatographed on normal phase SuperFlashTM Si column (sorbent mass 120 g and average particle size 40–60 m). Separation was carried out using a binary solvent system of ethyl acetate (mobile phase A) and methanol (mobile phase B) with a stepwise gradient starting from 10 to 90% methanol in 45 min at a flow rate of 20 mL/min. This procedure generated 45 fractions. A time scale of 1 min/fraction followed this time as the amount of extract loaded onto the column was low (0.85 g/run) in comparison to the polar extract (7.35 g/run). The UV absorptions of the eluates were monitored at the wavelengths of 280, 320, and 360 nm. All the above-mentioned fractions were analyzed for the antioxidant activity using DPPH and ferric ion reducing antioxidant power (FRAP) assays (Sections 2.4 and 2.5). The total phenolic content was determined by Folin–Ciocalteau method (Section 2.3). The highest three antioxidant fractions from each of the polar and nonpolar sets were selected for identification and quantification of the major polyphenols by LC–ESI-MS method (Section 2.6). The structure of the purified fraction of rosmarinic acid was further elucidated by NMR spectroscopy (Section 2.7).
2.3 Determination of total phenol (TP) The total phenolic content of marjoram extracts was determined using FCR as described by Singelton et al. [16]. The experiment was performed in two batches, which included three replications of each sample and standard. Methanolic gallic acid solutions (10–400 mg/L) were used as standards. In each replicate, 100 L of the appropriately diluted sample extract, 100 L methanol, 100 L FCR, and finally 700 L Na2 CO3 (20%) were added together and vortexed. The mixture was incubated for 20 min in the dark at room temperature. After incubation, the mixture was centrifuged at 13 000 rpm for 3 min. The absorbance of the supernatant was measured at 735 nm by UV-Vis spectrophotometry (Hitachi U-2900, Hitachi High-Technologies, Tokyo, Japan). The total phenolic content was expressed as mg gallic acid equivalent/100 g dry weight of the sample. www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 3205–3213
3207
Figure 1. Schematic diagram of antioxidant activity guided fractionation of marjoram extracts.
2.4 Determination of free radical scavenging activity by DPPH method A modified version of the DPPH assay with Trolox (a synthetic antioxidant) as a standard was used to measure in vitro antioxidant activity [17]. Briefly, 500 L of the extract or Trolox was added to 500 L of a methanol solution of DPPH (0.0476 mg/mL) in Eppendorf tubes. The Eppendorf tubes were incubated at room temperature for 30 min in the dark. The absorbance of the mixture was measured at 515 nm against the blank (methanol) using a UV-Vis spectrophotometer (Hitachi U-2900, Hitachi High-Technologies). C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
2.5 FRAP assay The FRAP assay was carried out as described by Stratil et al. [18] with slight modifications. The FRAP reagent was prepared by mixing 38 mM sodium acetate anhydrous in distilled water pH 3.6, 20 mM FeCl3 ·6H2 O in distilled water, and 10 mM 2,4,6-tri(2-pyridyl)-s-triazine in 40 mM HCl in a proportion of 10:1:1. This reagent was freshly prepared before each experiment. To each sample, 100 L of appropriately diluted sample extract and 900 L of FRAP reagent were added and the mixture was incubated at 37⬚C for 40 min in the dark. In the case of the blank, 100 L of methanol was
www.jss-journal.com
3208
J. Sep. Sci. 2014, 37, 3205–3213
M. B. Hossain et al.
Figure 2. Flash chromatographic fractionation of polar part of the marjoram extracts acquired in 280 nm (black line), 320 nm (red line), and 360 nm (blue line).
added to 900 L of FRAP reagent. The absorbance of the resulting solution was measured at 593 nm by spectrophotometry (Hitachi U-2900, Hitachi High-Technologies). Trolox at concentrations from 0.1 to 0.4 mM was used as a reference antioxidant standard. FRAP values were expressed as gram Trolox per 100 g dry weight of the sample.
2.6 LC–MS LC–MS analysis was performed on a Q-Tof Premier mass spectrometer coupled to Alliance 2695 HPLC system (Waters Corporation, Milford, MA, USA). The Q-Tof Premier is equipped with a lockspray source where an internal reference compound (Leucine–Enkephalin) was introduced simultaneously with the analyte for accurate mass measurements. Compounds were separated on an Atlantis T3 C18 column (100 × 2.1 mm; 3 m particle size) using 0.5% aqueous formic acid (solvent A) and 0.5% formic acid in 50:50 v/v acetonitrile/methanol (solvent B). The column temperature was maintained at 40⬚C. A stepwise gradient from 10 to 90% solvent B was applied at a flow rate of 0.2 mL/min for 26 min. Electrospray mass spectra data were recorded on a negative ionization mode for a mass range m/z 100– 1000. Capillary voltage and cone voltage were set at 3 kV and 30 V, respectively. Collision-induced dissociation (CID) of the analytes was achieved using 12–20 eV energy with argon as the collision gas.
pH 6.0 and 5 L of a 1% solution of TSP used as internal standard. Samples were then vortexed and transferred into a 5 mm NMR tube for 1 H NMR spectral data acquisition. 1 H NMR spectra were recorded on a BrukerTM Avance Spectrometer (Coventry, UK) at 500.162 MHz and 300 K using a 5 mm PABBO Broad Band Observed probe. Spectra were acquired using a water suppression pulse sequence (NOESYGPPR1D: RD-30⬚-t1-30⬚-tm-30⬚-acquire) set with a 4 s 30⬚ pulse, 10 ms mixing time (tm), 10 s relaxation delay, eight scans of 64 k data points were collected. The spectral width was 6009.615 Hz and the acquisition time was 5.45 s. The acquired spectra were line broadened (0.5 Hz), and manually phased using TOPSPIN v 1.3 software (Bruker) and baseline corrected. All spectra were then aligned with the TSP signal at ␦ = 0.00 ppm and resonance identification was performed by comparison of reference compounds acquired under similar conditions.
2.8 Statistical analysis Analysis of variance (ANOVA) was carried out using the software IBM SPSS Statistics version 19 (IBM, Sommers, New York). An ANOVA test was carried out for all experimental runs to determine significant differences among treatments at ␣ = 0.05 levels.
3 Results and discussion 2.7
1
H NMR spectroscopy of rosmarinic acid
Commercially available pure rosmarinic acid and dried flash fraction (no.17) containing only rosmarinic acid were resuspended in 0.6 mL of 400 mM deuterated phosphate buffer at C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
The initial extraction of the polyphenols of marjoram was carried out using 80% methanol. Aqueous methanol particularly 80% methanol has been shown to be a highly efficient extraction solvent for polyphenols from plant matrices [11,13]. This mixture of water and methanol is well capable of www.jss-journal.com
J. Sep. Sci. 2014, 37, 3205–3213
Liquid Chromatography
3209
Figure 3. Antioxidant activities of different flash fractions of marjoram (a) polar and (b) nonpolar extracts.
solubiizing both polar (predominantly phenolic acids) and nonpolar (predominantly flavonoids) polyphenols of plants. Therefore, liquid–liquid partitioning by water and ethyl acetate was carried out to separate the polar polyphenols from the nonpolar polyphenols, which also set a precedent to the subsequent flash chromatography. In a previous study, a total of 31 polyphenols distributed in four major categories; hydroxycinnamic acid derivatives, hydroxybenzoic acid derivatives, flavonoids, and phenolic terpenes have been reported [14]. However, no studies have been reported to associate the specific polyphenol(s) in marjoram with the antioxidant activity.
47 were considered as the main polar antioxidant polyphenols of marjoram. On the other hand, among the fractions of nonpolar part of the extract, fraction 7 had the highest TPC value. Two other discrete fractions namely fraction 12 and 17 also had high TPC. In fact, these fractions together with their adjacent fractions were responsible for 61% of the total TPC of nonpolar part of the marjoram extract. Therefore, fractions 7, 12, and 17 were selected for identifying the major nonpolar compounds of marjoram methanolic extract.
3.2 Antioxidant activity as measured by DPPH radical scavenging activity and FRAP 3.1 Total phenolic content The total phenol content (TPC) of the polar part of the methanolic primary extract was significantly (p < 0.05) higher than that of the nonpolar part. Similar results were also observed in the subsequent flash fractions where polar fractions in general had higher TPC than the nonpolar fractions. In fact, the average TPC of polar fractions was 64% higher than that of the nonpolar fractions. Among the polar fractions, fraction 33 had the highest TPC followed by fractions 34 and 32, all possibly containing the same compound. Other discrete fractions with considerable TPC were fractions 27 and 47, which were apparently due to the presence of different polyphenols to those in fraction 33. The adjacent fractions to fraction 47, i.e. fractions 46 and 48 also had high TPC values suggesting that their constituent polyphenols were similar. Since the fractionation was based on a time program of 0.5 min/fraction to avoid interference from the closely eluting compound, this resulted in a major peak fragmented into several fractions. The above-mentioned fractions (fractions 27, 33, and 47) collectively accounted for 43% of the total TPC of 90 fractions generated from polar part of the marjoram extract. For this reason, the polyphenols of fractions 27, 33, and C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Similar to the results of TPC, the antioxidant activity of polar part of the marjoram extract was significantly (p < 0.05) higher than that of nonpolar part (Fig. 3). The average FRAP and DPPH scavenging values of polar part were 126 and 92%, respectively, higher than those of nonpolar part (Fig. 3). There was a high correlation between DPPH scavenging and FRAP values with Pearson’s correlation coefficients (r) of 0.918 and 0.971 for polar and nonpolar part, respectively. This was also true for the relationship between antioxidant activity and TPC values. The Pearson’s correlation coefficients (r) between DPPH scavenging activity and TPC for polar and nonpolar parts were 0.871 and 0.973, respectively. The FRAP values were also closely associated with the TPC values showing the Pearson’s correlation coefficients (r) value of 0.926 and 0.929 for polar and nonpolar parts, respectively. The Pearson’s correlation coefficient values of the above-mentioned assays were in line with the high values of coefficients of determination with R2 ranging from 0.758 to 0.947. A high degree of correlation between antioxidant activity and TPC has also been reported by several authors [11, 13, 19]. This strongly indicated that polyphenols were the main antioxidant compounds in marjoram extracts. www.jss-journal.com
3210
J. Sep. Sci. 2014, 37, 3205–3213
M. B. Hossain et al.
Table 1. LC–MS/MS profile of the phenolic composition in the highest three flash chromatography fractions from polar and non-polar extracts
Fraction type Fraction number
RT (min) Phenolic compound
Empirical formula
Calculated [M−H]− (m/z)
Observed [M−H]− (m/z)
Major fragments [M−H]− (m/z)
Quantity (mg QEb) /Fr)
Polar
33 47
1.5 4.36 1.64 3.41 9.21 10.83
Caffeic acid glucoside Epigallocatechina) Arbutin Luteolin rutinoside Luteolin glucoronide Rosmarinic acida)
C15 H17 O9 − C15 H13 O7 − C12 H15 O7 − C27 H29 O15 − C21 H17 O12 − C18 H15 O8 −
341.0883 305.0665 271.0818 593.1533 461.0725 359.0763
341.0873 305.0661 271.0815 593.1506 461.0720 359.0767
30.65c) 46.60 18.81 10.57 23.91 14.34c)
7
7.38
Dihydroquercetin
C15 H11 O7 −
303.0505
303.0509
8
Dihydroluteolin
C15 H9 O6 −
287.0556
287.0551
Arbutin Apigenina) Quercetina)
C12 H15 O7 − C15 H9 O5 − C15 H9 O7 −
271.0818 269.0441 301.0334
271.0815 269.0450 301.0349
Gallocatechin derivative ND
ND
391.0654
8 10.17 10.46
Quercetin arabinoside C20 H17 O11 − Luteolin-7-O-glucosidea) C21 H19 O11 − Gallocatechin derivative ND
433.0771 447.0920 ND
433.0782 447.0927 414.0869
10.87
Rosmarinic acida)
359.0763
359.0767
179.0, 161.0, 135 289.0, 225.0 253.0, 108.0 285.0 285.0 197.0, 179.0, 161.0, 135.0 301.0, 285.0, 241.0, 227.0, 151.0, 135.0 227.0, 151.0, 135.0 253.0, 108.0 158.9 285.0, 227.1, 151.1, 135.0 305.0, 289.0, 225.0 301.0 285.0 305.0, 289.0, 225.0 197.0, 179.0, 161.0, 135.0
Nonpolar
27
10.17 10.46 7.38 12
17
7.38
C18 H15 O8 −
19.30
23.76 21.49 13.53 64.73 29.61 45.40 5.10 17.86 28.18c)
ND, not determined. a) Identification confirmed using commercial standards. b) Quercetin equivalent/fraction. c) Rosmarinic acid equivalent.
In the case of the polar part of the extract and in line with TPC data, fractions 27, 33, and 47 were the predominant antioxidant discrete fractions as measured by both DPPH scavenging and FRAP assays, while the fractions 7, 12, and 17 of the nonpolar part were the main antioxidant fractions accounting for 24% of the total DPPH scavenging values. In combination with the adjacent fractions, which possibly contained the same compound as the main fraction, this value increased to 53%. Similar results were also observed for FRAP values.
3.3 Identification and quantification of the polyphenols of the fractions with high antioxidant capacity The polyphenols of the antioxidant rich fractions were identified and quantified using LC–ESI-MS/MS. Polyphenols, in general, as they contain one or more hydroxyl and/or carboxylic acid groups, have higher sensitivity in negative ionization mode [20]. Therefore, data were acquired in negative ionization mode in the present study. Fourteen polyphenolic compounds were detected in the six selected C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
fractions with high antioxidant capacity (Fig. 1, Table 1). Therefore, these compounds could be considered as the principal antioxidants of marjoram extract. Identification of the five phenolic compounds (rosmarinic acid, epigallocatechin, quercetin, apigenin, luteolin-7-O-glucoside) was carried out by comparing retention times and their masses with those of the authentic standards. For the remaining nine compounds (caffeic acid hexoside, arbutin, luteolin rutinoside, luteolin glucoronide, dihydroquercetin, dihydroluteolin, quercetin arabinoside, and two gallocatechin derivatives) for which no standards were available, tentative identification was based on accurate mass measurements (observed mass error
J. Sep. Sci. 2014, 37, 3205–3213
Mohammad B. Hossain1 Gabriel Camphuis2 1 ´ Ingrid Aguilo-Aguayo Nirupama Gangopadhyay1 Dilip K. Rai1 1 Department
of Food Biosciences, Teagasc Food Research Centre, Ashtown, Dublin, Ireland 2 Polytechnique Institute of LaSalle, Beauvais, France Received May 30, 2014 Revised July 29, 2014 Accepted August 11, 2014
Research Article
Antioxidant activity guided separation of major polyphenols of marjoram (Origanum majorana L.) using flash chromatography and their identification by liquid chromatography coupled with electrospray ionization tandem mass spectrometry† Marjoram extracts have been separated into polar and nonpolar parts using liquid–liquid extraction. Both polar and nonpolar parts of the extracts were further fractionated by flash chromatography. The obtained fractions (90 polar and 45 nonpolar fractions) were investigated for their antioxidant activities by 2,2-diphenylpicrylhydrazyl and ferric ion reducing antioxidant power assays. A direct, positive, and linear relationship between antioxidant activity and total phenolic content of the fractions was observed. Based on antioxidant and total phenolic content data, the three fractions with the high antioxidant activities from polar and nonpolar part of the extract were analyzed for their constituent polyphenols by liquid chromatography coupled with electrospray ionization tandem mass spectrometry. Compounds were identified by matching the mass spectral data and retention time with those of authentic standards. Identification of the compounds for which there were no “in-house” standards available was carried out by accurate mass measurement of the precursor ions and product ions generated from collision-induced dissociation. Rosmarinic acid was found to be the strongest antioxidant polyphenol conferring the highest antioxidant activity to fractions 47 and 17 of polar and nonpolar part of the extract, respectively. The identification of the rosmarinic acid was further confirmed by 1 H NMR spectroscopy. Keywords: Antioxidant activity / Flash chromatography / Liquid–liquid partitioning / Polyphenols / Tandem mass spectrometry DOI 10.1002/jssc.201400597
1 Introduction In the past decades, scientific research interest has increased considerably in naturally occurring antioxidants for foods or medicinal applications since the use of synthetic antioxidants such as butylated hydroxyanisole and butylated hydroxytoluene is being restricted due to health and safety concerns [1, 2]. Natural antioxidants can protect the human body from free radicals and could retard the progress of many chronic diseases as well as lipid oxidative rancidity in foods [3–5]. A host of potentially beneficial physiological effects have Correspondence: Dr. Mohammad B. Hossain, Department of Food Biosciences, Teagasc Food Research Centre, Ashtown, Dublin 15, Ireland E-mail: [email protected] Fax: 00353-0-18059550
Abbreviations: CID, collision-induced dissociation; DPPH, 2,2-diphenylpicrylhydrazyl; FCR, Folin–Ciocalteu reagent; FRAP, ferric ion reducing antioxidant power; Trolox, (6hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid); TSP, 3-(trimethylsilyl)propoinic acid-d4
C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
been postulated for natural antioxidants over the past three decades, which are supported by extensive animal studies. Among these are beneficial influences on lipid metabolism, efficacy as antidiabetic, ability to stimulate digestion, antioxidant property, anticarcinogenic, and anti-inflammatory properties [6, 7]. Oxidation of polyunsaturated fatty acids not only lowers the nutritional value of food [8], but is also associated with cell membrane damage, aging, heart disease, and cancer in living organisms [9]. Therefore, the addition of natural antioxidants to food products has become popular as means of increasing shelf life and to reduce wastage and nutritional losses by inhibiting and delaying oxidation [10]. Several studies reported that extracts of Origanum majorana (marjoram) had high antioxidant capacity [11–13] mostly due to the polyphenolic compounds present in them. A total of 31 polyphenols
†This paper is included in the virtual special issue sample preparation in mass spectrometry available at the Journal of Separation Science website. Colour Online: See the article online to view Fig. 3 in colour.
www.jss-journal.com
3206
M. B. Hossain et al.
were identified in marjoram in a previous study [14]. However, to the best of our knowledge, no study has demonstrated which of these polyphenols of marjoram were mainly responsible for its high antioxidant activity. Therefore, an antioxidant activity guided fractionation approach is followed to establish this relationship in the present study. A rapid and easyto-use solvent-based partitioning in combination with automated flash chromatographic fractionation has been used to separate the most antioxidant fractions. The importance of these sample preparation techniques and various other sample cleanup methods for analyses of polyphenols has been extensively reviewed by Stalikas [15]. Subsequent structural characterization and quantification of major polyphenols in Origanum majorana by LC–ESI-MS/MS have been presented.
2 Materials and methods 2.1 Samples and reagents Dried and ground marjoram was provided by AllinAll Ingredients, Dublin, Ireland. According to product specifications, the country of origin of the spice used was Turkey. The spice was air-dried after heat treatment (steam sterilization at 120⬚C for 30 s). The dried spice was ground (particle size range 500–600 m) and stored at −20⬚C in darkness. Three polyphenol standards namely epigallocatechin, quercetin, and rosmarinic acid, and an NMR internal reference 3-(trimethylsilyl)propoinic acid-d4 (TSP) sodium salt were purchased from Sigma–Aldrich, Wicklow, Ireland. Two flavonoid standards, apigenin and luteolin-7O-glucoside, were purchased from Extrasynthese, France. HPLC-grade methanol, acetonitrile, ethylacetate, and water were purchased from VWR International, Leicestershire, UK and Lennox Laboratory Supplies, Dublin, Ireland, respectively. The purity of standards and solvents were in the range of 95–99.8%. Folin–Ciocalteu reagent (FCR), gallic acid, sodium carbonate (Na2 CO3 ), hydrochloric acid (HCl), formic acid, deuterated phosphate buffer, 2,2-diphenylpicrylhydrazyl (DPPH), sodium acetate anhydrous, ferric chloride hexahydrate, 2,4,6-tri(2-pyridyl)-s-triazine, and 6-hydroxy-2,5,7,8tetramethylchroman-2-carboxylic acid (Trolox) were purchased from Sigma–Aldrich. 2.2 Preparation of solid/liquid extract and its fractionation Antioxidant-guided fractionation of marjoram extracts was carried out as shown in Fig. 1. Dried and ground spice samples (50 g) were sequentially extracted twice using 500 mL of 80% aqueous methanol, each time in the dark at room temperature (23⬚C). The sample suspension was shaken overnight in an orbital shaker (MaxQ 6000 Shaker, Thermo Fisher Scientific, MA, USA) set at 150 rpm at room temperature. The mixture was then filtered through a Buchner funnel (1 m). The extracts were dried immediately under reduced pressure using a rotary evaporator (Heidolph C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Sep. Sci. 2014, 37, 3205–3213
Rotary Evaporator, Germany) with a water bath set at 50⬚C. This dried extract was redissolved in water (500 mL) and partitioned against ethylacetate (500 mL). The aqueous polar part contained 89.6% w/w of the methanolic extract. This extract was freeze-dried and resuspended in a minimal amount of water (60 mL), which was further fractionated using RP flash chromatography (Varian IntelliFlash 310, CA, USA). Flash chromatography was performed on a SuperFlashTM C18 column with a sorbent mass of 300 g and average size of the particles was 40–60 m. A stepwise gradient from 10% aqueous methanol to 90% methanol in 45 min at a flow rate of 40 mL/min was used to separate the polyphenols of the polar extracts (Fig. 2). A total of 90 fractions were generated following a time scale of 0.5 min/fraction. Meanwhile the nonpolar ethylacetate extract was dried under reduced pressure as mentioned for methanolic extract. The dried nonpolar extract was also redissolved in minimal amount of fresh ethyl acetate (50 mL) and chromatographed on normal phase SuperFlashTM Si column (sorbent mass 120 g and average particle size 40–60 m). Separation was carried out using a binary solvent system of ethyl acetate (mobile phase A) and methanol (mobile phase B) with a stepwise gradient starting from 10 to 90% methanol in 45 min at a flow rate of 20 mL/min. This procedure generated 45 fractions. A time scale of 1 min/fraction followed this time as the amount of extract loaded onto the column was low (0.85 g/run) in comparison to the polar extract (7.35 g/run). The UV absorptions of the eluates were monitored at the wavelengths of 280, 320, and 360 nm. All the above-mentioned fractions were analyzed for the antioxidant activity using DPPH and ferric ion reducing antioxidant power (FRAP) assays (Sections 2.4 and 2.5). The total phenolic content was determined by Folin–Ciocalteau method (Section 2.3). The highest three antioxidant fractions from each of the polar and nonpolar sets were selected for identification and quantification of the major polyphenols by LC–ESI-MS method (Section 2.6). The structure of the purified fraction of rosmarinic acid was further elucidated by NMR spectroscopy (Section 2.7).
2.3 Determination of total phenol (TP) The total phenolic content of marjoram extracts was determined using FCR as described by Singelton et al. [16]. The experiment was performed in two batches, which included three replications of each sample and standard. Methanolic gallic acid solutions (10–400 mg/L) were used as standards. In each replicate, 100 L of the appropriately diluted sample extract, 100 L methanol, 100 L FCR, and finally 700 L Na2 CO3 (20%) were added together and vortexed. The mixture was incubated for 20 min in the dark at room temperature. After incubation, the mixture was centrifuged at 13 000 rpm for 3 min. The absorbance of the supernatant was measured at 735 nm by UV-Vis spectrophotometry (Hitachi U-2900, Hitachi High-Technologies, Tokyo, Japan). The total phenolic content was expressed as mg gallic acid equivalent/100 g dry weight of the sample. www.jss-journal.com
Liquid Chromatography
J. Sep. Sci. 2014, 37, 3205–3213
3207
Figure 1. Schematic diagram of antioxidant activity guided fractionation of marjoram extracts.
2.4 Determination of free radical scavenging activity by DPPH method A modified version of the DPPH assay with Trolox (a synthetic antioxidant) as a standard was used to measure in vitro antioxidant activity [17]. Briefly, 500 L of the extract or Trolox was added to 500 L of a methanol solution of DPPH (0.0476 mg/mL) in Eppendorf tubes. The Eppendorf tubes were incubated at room temperature for 30 min in the dark. The absorbance of the mixture was measured at 515 nm against the blank (methanol) using a UV-Vis spectrophotometer (Hitachi U-2900, Hitachi High-Technologies). C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
2.5 FRAP assay The FRAP assay was carried out as described by Stratil et al. [18] with slight modifications. The FRAP reagent was prepared by mixing 38 mM sodium acetate anhydrous in distilled water pH 3.6, 20 mM FeCl3 ·6H2 O in distilled water, and 10 mM 2,4,6-tri(2-pyridyl)-s-triazine in 40 mM HCl in a proportion of 10:1:1. This reagent was freshly prepared before each experiment. To each sample, 100 L of appropriately diluted sample extract and 900 L of FRAP reagent were added and the mixture was incubated at 37⬚C for 40 min in the dark. In the case of the blank, 100 L of methanol was
www.jss-journal.com
3208
J. Sep. Sci. 2014, 37, 3205–3213
M. B. Hossain et al.
Figure 2. Flash chromatographic fractionation of polar part of the marjoram extracts acquired in 280 nm (black line), 320 nm (red line), and 360 nm (blue line).
added to 900 L of FRAP reagent. The absorbance of the resulting solution was measured at 593 nm by spectrophotometry (Hitachi U-2900, Hitachi High-Technologies). Trolox at concentrations from 0.1 to 0.4 mM was used as a reference antioxidant standard. FRAP values were expressed as gram Trolox per 100 g dry weight of the sample.
2.6 LC–MS LC–MS analysis was performed on a Q-Tof Premier mass spectrometer coupled to Alliance 2695 HPLC system (Waters Corporation, Milford, MA, USA). The Q-Tof Premier is equipped with a lockspray source where an internal reference compound (Leucine–Enkephalin) was introduced simultaneously with the analyte for accurate mass measurements. Compounds were separated on an Atlantis T3 C18 column (100 × 2.1 mm; 3 m particle size) using 0.5% aqueous formic acid (solvent A) and 0.5% formic acid in 50:50 v/v acetonitrile/methanol (solvent B). The column temperature was maintained at 40⬚C. A stepwise gradient from 10 to 90% solvent B was applied at a flow rate of 0.2 mL/min for 26 min. Electrospray mass spectra data were recorded on a negative ionization mode for a mass range m/z 100– 1000. Capillary voltage and cone voltage were set at 3 kV and 30 V, respectively. Collision-induced dissociation (CID) of the analytes was achieved using 12–20 eV energy with argon as the collision gas.
pH 6.0 and 5 L of a 1% solution of TSP used as internal standard. Samples were then vortexed and transferred into a 5 mm NMR tube for 1 H NMR spectral data acquisition. 1 H NMR spectra were recorded on a BrukerTM Avance Spectrometer (Coventry, UK) at 500.162 MHz and 300 K using a 5 mm PABBO Broad Band Observed probe. Spectra were acquired using a water suppression pulse sequence (NOESYGPPR1D: RD-30⬚-t1-30⬚-tm-30⬚-acquire) set with a 4 s 30⬚ pulse, 10 ms mixing time (tm), 10 s relaxation delay, eight scans of 64 k data points were collected. The spectral width was 6009.615 Hz and the acquisition time was 5.45 s. The acquired spectra were line broadened (0.5 Hz), and manually phased using TOPSPIN v 1.3 software (Bruker) and baseline corrected. All spectra were then aligned with the TSP signal at ␦ = 0.00 ppm and resonance identification was performed by comparison of reference compounds acquired under similar conditions.
2.8 Statistical analysis Analysis of variance (ANOVA) was carried out using the software IBM SPSS Statistics version 19 (IBM, Sommers, New York). An ANOVA test was carried out for all experimental runs to determine significant differences among treatments at ␣ = 0.05 levels.
3 Results and discussion 2.7
1
H NMR spectroscopy of rosmarinic acid
Commercially available pure rosmarinic acid and dried flash fraction (no.17) containing only rosmarinic acid were resuspended in 0.6 mL of 400 mM deuterated phosphate buffer at C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
The initial extraction of the polyphenols of marjoram was carried out using 80% methanol. Aqueous methanol particularly 80% methanol has been shown to be a highly efficient extraction solvent for polyphenols from plant matrices [11,13]. This mixture of water and methanol is well capable of www.jss-journal.com
J. Sep. Sci. 2014, 37, 3205–3213
Liquid Chromatography
3209
Figure 3. Antioxidant activities of different flash fractions of marjoram (a) polar and (b) nonpolar extracts.
solubiizing both polar (predominantly phenolic acids) and nonpolar (predominantly flavonoids) polyphenols of plants. Therefore, liquid–liquid partitioning by water and ethyl acetate was carried out to separate the polar polyphenols from the nonpolar polyphenols, which also set a precedent to the subsequent flash chromatography. In a previous study, a total of 31 polyphenols distributed in four major categories; hydroxycinnamic acid derivatives, hydroxybenzoic acid derivatives, flavonoids, and phenolic terpenes have been reported [14]. However, no studies have been reported to associate the specific polyphenol(s) in marjoram with the antioxidant activity.
47 were considered as the main polar antioxidant polyphenols of marjoram. On the other hand, among the fractions of nonpolar part of the extract, fraction 7 had the highest TPC value. Two other discrete fractions namely fraction 12 and 17 also had high TPC. In fact, these fractions together with their adjacent fractions were responsible for 61% of the total TPC of nonpolar part of the marjoram extract. Therefore, fractions 7, 12, and 17 were selected for identifying the major nonpolar compounds of marjoram methanolic extract.
3.2 Antioxidant activity as measured by DPPH radical scavenging activity and FRAP 3.1 Total phenolic content The total phenol content (TPC) of the polar part of the methanolic primary extract was significantly (p < 0.05) higher than that of the nonpolar part. Similar results were also observed in the subsequent flash fractions where polar fractions in general had higher TPC than the nonpolar fractions. In fact, the average TPC of polar fractions was 64% higher than that of the nonpolar fractions. Among the polar fractions, fraction 33 had the highest TPC followed by fractions 34 and 32, all possibly containing the same compound. Other discrete fractions with considerable TPC were fractions 27 and 47, which were apparently due to the presence of different polyphenols to those in fraction 33. The adjacent fractions to fraction 47, i.e. fractions 46 and 48 also had high TPC values suggesting that their constituent polyphenols were similar. Since the fractionation was based on a time program of 0.5 min/fraction to avoid interference from the closely eluting compound, this resulted in a major peak fragmented into several fractions. The above-mentioned fractions (fractions 27, 33, and 47) collectively accounted for 43% of the total TPC of 90 fractions generated from polar part of the marjoram extract. For this reason, the polyphenols of fractions 27, 33, and C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Similar to the results of TPC, the antioxidant activity of polar part of the marjoram extract was significantly (p < 0.05) higher than that of nonpolar part (Fig. 3). The average FRAP and DPPH scavenging values of polar part were 126 and 92%, respectively, higher than those of nonpolar part (Fig. 3). There was a high correlation between DPPH scavenging and FRAP values with Pearson’s correlation coefficients (r) of 0.918 and 0.971 for polar and nonpolar part, respectively. This was also true for the relationship between antioxidant activity and TPC values. The Pearson’s correlation coefficients (r) between DPPH scavenging activity and TPC for polar and nonpolar parts were 0.871 and 0.973, respectively. The FRAP values were also closely associated with the TPC values showing the Pearson’s correlation coefficients (r) value of 0.926 and 0.929 for polar and nonpolar parts, respectively. The Pearson’s correlation coefficient values of the above-mentioned assays were in line with the high values of coefficients of determination with R2 ranging from 0.758 to 0.947. A high degree of correlation between antioxidant activity and TPC has also been reported by several authors [11, 13, 19]. This strongly indicated that polyphenols were the main antioxidant compounds in marjoram extracts. www.jss-journal.com
3210
J. Sep. Sci. 2014, 37, 3205–3213
M. B. Hossain et al.
Table 1. LC–MS/MS profile of the phenolic composition in the highest three flash chromatography fractions from polar and non-polar extracts
Fraction type Fraction number
RT (min) Phenolic compound
Empirical formula
Calculated [M−H]− (m/z)
Observed [M−H]− (m/z)
Major fragments [M−H]− (m/z)
Quantity (mg QEb) /Fr)
Polar
33 47
1.5 4.36 1.64 3.41 9.21 10.83
Caffeic acid glucoside Epigallocatechina) Arbutin Luteolin rutinoside Luteolin glucoronide Rosmarinic acida)
C15 H17 O9 − C15 H13 O7 − C12 H15 O7 − C27 H29 O15 − C21 H17 O12 − C18 H15 O8 −
341.0883 305.0665 271.0818 593.1533 461.0725 359.0763
341.0873 305.0661 271.0815 593.1506 461.0720 359.0767
30.65c) 46.60 18.81 10.57 23.91 14.34c)
7
7.38
Dihydroquercetin
C15 H11 O7 −
303.0505
303.0509
8
Dihydroluteolin
C15 H9 O6 −
287.0556
287.0551
Arbutin Apigenina) Quercetina)
C12 H15 O7 − C15 H9 O5 − C15 H9 O7 −
271.0818 269.0441 301.0334
271.0815 269.0450 301.0349
Gallocatechin derivative ND
ND
391.0654
8 10.17 10.46
Quercetin arabinoside C20 H17 O11 − Luteolin-7-O-glucosidea) C21 H19 O11 − Gallocatechin derivative ND
433.0771 447.0920 ND
433.0782 447.0927 414.0869
10.87
Rosmarinic acida)
359.0763
359.0767
179.0, 161.0, 135 289.0, 225.0 253.0, 108.0 285.0 285.0 197.0, 179.0, 161.0, 135.0 301.0, 285.0, 241.0, 227.0, 151.0, 135.0 227.0, 151.0, 135.0 253.0, 108.0 158.9 285.0, 227.1, 151.1, 135.0 305.0, 289.0, 225.0 301.0 285.0 305.0, 289.0, 225.0 197.0, 179.0, 161.0, 135.0
Nonpolar
27
10.17 10.46 7.38 12
17
7.38
C18 H15 O8 −
19.30
23.76 21.49 13.53 64.73 29.61 45.40 5.10 17.86 28.18c)
ND, not determined. a) Identification confirmed using commercial standards. b) Quercetin equivalent/fraction. c) Rosmarinic acid equivalent.
In the case of the polar part of the extract and in line with TPC data, fractions 27, 33, and 47 were the predominant antioxidant discrete fractions as measured by both DPPH scavenging and FRAP assays, while the fractions 7, 12, and 17 of the nonpolar part were the main antioxidant fractions accounting for 24% of the total DPPH scavenging values. In combination with the adjacent fractions, which possibly contained the same compound as the main fraction, this value increased to 53%. Similar results were also observed for FRAP values.
3.3 Identification and quantification of the polyphenols of the fractions with high antioxidant capacity The polyphenols of the antioxidant rich fractions were identified and quantified using LC–ESI-MS/MS. Polyphenols, in general, as they contain one or more hydroxyl and/or carboxylic acid groups, have higher sensitivity in negative ionization mode [20]. Therefore, data were acquired in negative ionization mode in the present study. Fourteen polyphenolic compounds were detected in the six selected C 2014 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
fractions with high antioxidant capacity (Fig. 1, Table 1). Therefore, these compounds could be considered as the principal antioxidants of marjoram extract. Identification of the five phenolic compounds (rosmarinic acid, epigallocatechin, quercetin, apigenin, luteolin-7-O-glucoside) was carried out by comparing retention times and their masses with those of the authentic standards. For the remaining nine compounds (caffeic acid hexoside, arbutin, luteolin rutinoside, luteolin glucoronide, dihydroquercetin, dihydroluteolin, quercetin arabinoside, and two gallocatechin derivatives) for which no standards were available, tentative identification was based on accurate mass measurements (observed mass error
