Article pubs.acs.org/ac
Development of a Signal-Ratio-Based Antioxidant Index for Assisting the Identification of Polyphenolic Compounds by Mass Spectrometry Che-I Liao and Kuo-Lung Ku* Department of Applied Chemistry, National Chiayi University, Chiayi City 60004, Taiwan ABSTRACT: A new concept, called the signal-ratio-based antioxidant index (SRBAI), is proposed for the identification of antioxidants. The SRBAI is derived from the signal ratio of free-radical scavenging to polyphenolic chromophore absorbance. Each SRBAI value corresponds to a specific antioxidant under the same antioxidant assay condition. Hence, the SRBAI can be used as an identification card of the antioxidant, which can resolve even components with the same retention time or with similar mass fragmentation spectra. We employed onion and several quercetin glycosides as models. There are four major peaks of onion with free-radical scavenging ability. One of the peaks failed to be identified; for the authentic compounds, quercetion-3-glucoside (SRBAI = 8.98) and quercetin-3-rutinoside (SRBAI = 3.01) eluted at the same retention time as the unknown peak. However, the unknown was considered to be quercetin-3-glucoside for its SRBAI was 8.91. SRBAI values also can be applied to differentiate the unknown peaks with the same parent ion of m/z 463 and collision-induced dissociation (CID) spectra as the two authentic compounds, quercetin-3-glucoside and quercetin-4′-glucoside. This newly introduced SRBAI can act as an efficient and precise identification tool, especially for online identification of similar polyphenolic isomers. olyphenolic compounds or “phenolics” are a complex group of compounds of plant origin. The basic structural characteristic of all polyphenolics is the presence of one or more hydroxylated benzene rings. The two main groups of polyphenolics are phenolic acids and flavonoids. Phenolic acids can be further classified into (1) benzoic acids and their derivatives or (2) and cinnamic acids and their derivatives. The main subclasses of flavonoids are flavanols (e.g., catechin and related compounds), flavonols (e.g., quercetin and related glycosides), flavones (e.g., apigenin and luteolin), flavanones (e.g., naringin), and isoflavones (e.g., genistein and daidzein). All of these occur in nature mostly as glycosides. More than 4000 phenolic acids and 5000 flavonoids are currently known, which indicates the chemical complexity of these compounds.1 Interest in phenolic acids and flavonoids is continuously increasing, because of their antioxidant properties, among other characteristics. In recent years, high-performance liquid chromatography (HPLC) combined with diode-array detection and mass spectrometry (MS) have proved to be powerful techniques for the separation and identification of polyphenolics in complex samples.1 Antioxidant properties have been determined by the radical scavenging activity,2 such as that of 2,2-diphenyl-1-picrylhydrazyl (DPPH)3−10 or 2,2′-azinobis(3ethylbenzothiazoline-6-sulfonic acid (ABTS).11−17 Meanwhile, the complete identification of phenolic acids and flavonoids is impossible, because they have high chemical complexity18 and their extraction from plant matrices is difficult; moreover, the analysis of these compounds in food samples is affected by complex interference from nonphenolic
P
© 2012 American Chemical Society
compounds and nonflavonoids, and there are very few commercially available standards for these compounds. Furthermore, even using the retention times of authentic compounds for identification, problems remain. First, in real samples, polyphenolic compounds have similar polarities and structures. If different compounds elute with the same retention times, the absorption spectra from the diode-array detector might also be similar,19 rendering identification of the authentic compounds difficult. Second, even if mass spectra are alternatively provided, the time-consuming manual analysis of these data is nontrivial, because the fragmentation of small molecules under varying fragmentation energies is not completely understood.20,21 Furthermore, because of the limited reproducibility of collision-induced dissociation (CID) mass spectra on different instruments, even searching in spectral libraries is a serious problem.22−24 According to these problems, implying that accurate identification is very difficult. Therefore, we propose a new conceptthe signal-ratiobased antioxidant index (SRBAI)for the characterization of antioxidant components and determination of the antioxidant capacity. To derive the SRBAI, antioxidant-containing mixtures are first separated by HPLC; the eluents are then directed online to the first detector to measure the absorbance at 280 nm and to the second detector to measure antioxidant activity Received: May 27, 2012 Accepted: August 9, 2012 Published: August 9, 2012 7440
dx.doi.org/10.1021/ac301283s | Anal. Chem. 2012, 84, 7440−7448
Analytical Chemistry
Article
Figure 1. Instrumental setup of the signal-ratio-based antioxidant index (SRBAI) with an HPLC-ABTS• + online radical-scavenging system.
Figure 2. Diagram of the basic concept of SRBAI.
hydroquinone, resorcinol, catechol, trolox, rosamaric acid, quercetin-3-β-D-glucoside, rutin, quercetin-4′-O-β-glucoside, and quercetin-3,4′-di-O-β-glucoside were prepared in methanol and stored at −20 °C. Antioxidant standard solutions in methanol were prepared daily from the stock solutions in 7-mL dark vials. The ammonium acetate buffer (pH 6.0) used in the real sample analysis contained 10 mM H3CCOONH4/H3CCOOH (Merck, Darmstadt, Germany) in water. Chemical Reagents. The ABTS solution contained 0.5 mM ABTS, 50 μM K2HPO4, and 1.75 mM K2S2O8 in 0.5 L double-distilled (dd) H2O, and it was stored in darkness for 12−16 h. HPLC−ABTS Conditions. A block schematic of the instrumental setup is presented in Figure 1. The HPLC system consisted of the following: an HPLC eluent pump (Hitachi Model L7100, Japan), a UV−vis detector (Shimadzu Model SPD-10a, Japan), and a UV−vis absorbance detector (Hitachi Model L7420, Japan) equipped with a tungsten lamp. Separations were carried out on a Model Mightysil RP-18 column (250 mm × 4.6 mm; inner diameter (id) = 5 μm; Kanto Chemical Co., Inc., Tokyo, Japan). Samples were injected using a Model 7725 injector fitted with a 20-μL
signals. Then, the SRBAI is derived for each individual component from the ratio of the antioxidant activity signal to the absorbance at 280 nm. In this study, we tested the SRBAIs of several polyphenolic compounds with different concentrations and investigated the effects of mobile phase composition, free-radical reagent concentration, and free-radical reagent buffer concentration.
■
EXPERIMENTAL SECTION Chemicals and Solvents. Quercetin dehydrate, (+)-catechin dehydrate, epicatechin, kaempferol, caffeic acid, cyanidin, trolox, rosamaric acid, quercetin-3-β-D-glucoside, quercetin-3rutinoside (rutin), 2,2-azinobis-(3-ethylbenzothiazoline-6-sulfonate) (ABTS), potassium phosphate dibasic, and potassium persulfate were obtained from Sigma−Aldrich (St. Louis, MO, USA). Quercetin-4′-O-β-glucoside and quercetin-3,4′-di-O-βglucoside were obtained from Polyphenols Laboratories AS, (Sandnes, Norway). HPLC-grade solvents were purchased from Merek (Darmstadt, Germany). Ultrapure water was obtained from a Milli-Q system (Millipore Corp., Marlborough, MA). Preparation of Solutions. Antioxidant stock solutions of quercetin dehydrate, (+)-catechin dehydrate, epicatechin, 7441
dx.doi.org/10.1021/ac301283s | Anal. Chem. 2012, 84, 7440−7448
Analytical Chemistry
Article
Demonstration of the Concept. In the present study, nine polyphenolic compounds were used as model compounds, including three phenolic acids (caffeic acid, rosmaric acid, and trolox) and six flavonoids (cyanidin, catechin, epicatechin, quercetin, kaempferol, and quercetin-3,4′-diglucoside (Q-3,4′G)). The SRBAI values listed in Table 1 were derived from the ratio of the extinction area of the free radical to the UV absorbance. Theoretically, the SRBAI is not only specific for each compound but also constant across different concentrations. We carried out the test at five different concentrations: 5, 10, 50, 100, and 200 μM. The SRBAI at the lowest concentration (5 μM) was lower than that at 10 or 50 μM for each compound, except in the case of cyanidin. At the other extreme, the SRBAI at the highest concentration (200 μM) was lower than other concentrations in the case of Q-3,4′-G. Theoretically, SRBAI values will remain constant over a concentration range, which is the overlapping range for both the responding linear of scavenging extinction signal and absorbance. If the concentrations of sample exceeded any one of the linear responding range, the SRBAI would be changed and unaccepted. Hence, determining the linearity between the sample concentration and the responding signal is important. Here, we employed a coefficient of determination (r2) to evaluate the linearity and set the r2 value of any acceptable linearity to be equal to or larger than 0.995 for five concentration points. Hence, any range with a linearity of
