Phenolic Compounds and Antioxidant Activities of Liriope muscari

Five phenolic compounds, namely N-trans-coumaroyltyramine (1), N-trans-feruloyltyramine (2), N-trans-feruloyloctopamine (3), 5,7-dihydroxy-8-methoxyflavone (4) and (3S)3,5,4′-trihydroxy-7-methoxy-6-methylhomoisoflavanone (5), were isolated from the fibrous roots of Liriope muscari (Liliaceae). Compounds 2–5 were isolated for the first time from the Liriope genus. Their in vitro antioxidant activities were assessed by the DPPH and ABTS scavenging methods with microplate assays. The structure-activity relationships of compounds 1–3 are discussed.


Introduction
Antioxidant activity usually means the ability of a compound to delay, inhibit, or prevent the oxidation of oxidizable materials by scavenging free radicals and reducing oxidative stress [1]. Antioxidants can scavenge ROS (reactive oxygen species) to protect the cells from damage caused by the latter. At present, the most commonly used antioxidants include vitamin C (VC), vitamin E, butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT), propyl gallate and tert-butyl OPEN ACCESS hydroquinone. However, some chemically synthesized antioxidants like BHA and BHT are now being restricted by legislation because of doubts over their possible toxic and carcinogenic effects [2]. Therefore, there is a growing interest in finding antioxidants from natural sources [3,4]. Assays based on the use of 1,1-diphenyl-2-picryl-hydrazyl (DPPH) and 2,2′-azino-bis(3-ethylbenzthiazoline-6sulfonic acid) (ABTS) radicals are among the most commonly used spectrophotometric methods for determination of the antioxidant capacity of foods, beverages, plant extracts and pure compounds due to the simple, rapid, sensitive, and reproducible procedures involved [5,6]. Phenolic compounds usually possess different antioxidant activity potentials because of their phenolic hydroxy groups which can act as a hydrogen or electron donor [7]. Phenolic acids, flavonoids and tannins are wellknown potential natural antioxidants [1]. The hunt for effective and safe antioxidants from natural products is considered to be a shortcut [8].

OH
To our knowledge, this is the first time that the phenolic components of L. muscari have been studied. Their in vitro antioxidant activities were assessed by the DPPH and ABTS scavenging method. N-trans-feruloyltyramine (IC 50 28.7, 8.2 μg/mL) and N-trans-feruloyloctopamine (IC 50 14.4, 7.6 μg/mL) showed potential antioxidant activities.

Isolation and Characterization of Compounds 1-5
The compounds were isolated using silica gel and Sephadex LH-20 gel column chromatography from 80% ethanol extract of L. muscari. The structures of compounds 1-4 were characterized by examination of their ESI-MS, NMR (1H-and 13C-) data and comparison with literature reports.
Compound 5 was first isolated in 1985 from Ophiopogonis [14] (Liliaceae), a closely linked genus that can be easily confused with Liriope. In the original paper, the planar structure of compound 5 was identified by comparing the 1 H-NMR data with that of (3S)-3,5,7-trihydroxy-4′-methoxy homoisoflavonone (eucomol, a typical homoisoflavonone isolated from Eucomis bicolor BAK. (Liliaceae) [15,16]). In our study, the structure was further confirmed using 13 C-NMR, APT and 2D-NMR techniques, including 1 H-1 HCOSY, HSQC, HMBC ( Figure 2). The configuration at C-3 was determined to be (S), similar to that of eucomol, based on the positive sign of its specific rotation [16,17]. For a long time, it was believed that there were no homoisoflavones in Liriope, so this is the first time a homoisoflavone has been isolated from the Liriope genus. The 1,1-diphenyl-2-picrylhydrazyl radical (DPPH), which possesses an unpaired electron and exhibits a stable violet color in methanol solution (peak absorbance at 517 nm), is commonly used as a reagent for evaluation of the free radical scavenging activity of antioxidants [18]. The DPPH assay is based on the reduction of DPPH in methanol solution in the presence of a hydrogen-donating antioxidant due to the formation of the non-radical form (DPPH-H) in the reaction [19]. Figure 3 shows the DPPH scavenging activities of compounds 1-5 and reference antioxidants at different concentrations (12.5-100 μg/mL). The test compounds exhibited different DPPH scavenging activities in a concentration-dependent manner. The scavenging effects of compounds 1-5 and reference antioxidants on DPPH decreased in the following order: VC > compound 3 > compound 2 > BHT > compound 4 > compound 1 > compound 5. The inhibition ratios at a concentration of 25 μg/mL are listed in Table 1. Compounds 2 and 3 exhibited effective radical scavenging activity while compounds 1, 4, 5 showed very weak activity.

ABTS Scavenging Activity
In this assay, the 2,2′-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) radical, which has a peak absorbance at 734 nm, should be preformed by mixing ABTS and potassium persulfate (K 2 S 2 O 8 ). When antioxidants were added, the ABTS radical, which has a blue-green color, is reduced to ABTS (no color). Different decoloration abilities indicate different ABTS scavenging activities [20]. Figure 4 shows the ABTS scavenging abilities of compounds 1-5 and reference standards. The test compounds also exhibited different radical scavenging activities in a concentration-dependent manner like in the DPPH assay. The scavenging effects of compounds 1-5 and reference antioxidants on ABTS ·+ decreased in the following order: VC ≈ BHT > compound 2 > compound 3 > compound 1 > compound 4 > compound 5. The inhibition ratios at concentration of 25 μg/mL are listed in Table 1. Compounds 2 and 3 exhibited effective radical scavenging activity, while compounds 1, 4 and 5 showed relatively weak activity. In both methods, compounds 2 [21] and 3 [22] showed potential activity, while compounds 1 [23,24] and 4 [25] showed very weak antioxidant activity, which is consistent with the reported results. Furthermore, for compound 4, quantitative structure-activity relationship analysis [26] also suggested it would not show effective activity because of the absence of 3-OH and o-dihydroxy structure in the B ring, and less free -OH groups (only two) in the structure, which are required for high antioxidant activity. As for compound 5, its antioxidant activity has been previously evaluated using an on-line HPLC-DAD-CL method based on hydrogen peroxide elimination [27]. However, the results cannot be compared with our data due to the absence of any mention of the concentration used in that study.
It is interesting to investigate the structure-activity relationship for compounds 1-3. These compounds have similar structures, but very different activities. Comparing their structures (Figure 1), the main differences are the substituents at C-3 (R 1 ) and C-7′ (R 2 ). Comparing the structures and activities of compounds 2 and 3 (p > 0.05, 25 μg/mL) it is inferred that the presence of methyl group at C-7′ seems to have some, but little influence on antioxidant activity. By comparing compounds 1 and 2 (p < 0.05, 25 μg/mL), it is inferred that the presence of methyl group at C-3 is the key factor that affects the activities, therefore, a radical scavenging activity mechanism represented by the reaction shown in Figure 5 is proposed, using DPPH as an example. Compounds 2 and 3 have two resonance structures A and B stabilizing the product, thereby exhibiting superior antioxidant activity than compound 1. As HPLC evaluation (Experimental section) indicates a relative purity between 90.1% and 96.7% for compounds 1-3, potential synergistic effects from minor impurities could also contribute to the observed activity.  All other chemicals were of analytical reagent grade and used without any further purification.

Plant Material
Fresh fibrous roots of L. muscari were collected from Quanzhou City, Fujian Province, China, in May 2010. The species was identified by Dr. Zhang. J. (National Institutes for Food and Drug Control, NIFDC for short). The voucher specimens were deposited at the herbarium of NIFDC. The roots were air-dried and ground to a powder using a grinding mill (Retsch Muhle, Haan, Germany).

Compound Isolation
In this part, there were mainly two procedures including enrichment and isolation. For the enrichment of phenolic compounds, polyamide resin was used. Polyamide is a commonly used stationary phase for the isolation of phenolic compounds. The enriching mechanism is based on the adsorption power from hydrogen bonds between carbonyl groups of the polyamide and the phenolic hydroxyl groups of target compounds or the polyamide amides and carbonyl groups of fatty acids, etc., The strength of adsorption is dependent on the number of phenolic hydroxyls exposed and their position in the molecule [28] and the power is strongest in water, while getting weaker when the ethanol concentration in the mobile phase is increased.
The detailed procedures are as follows: the powder (2 kg) was extracted three times with 80% hot ethanol (1 L), for 1 h each time. The extracts were concentrated to afford a syrup (1 kg), which was dissolved in 10% ethanol (4 L). Polyamide (1 kg) was added into the solution and stirred about 1 h to make sure the phenolic compounds were adsorbed on the polyamide to some extent. Then the polyamide was centrifuged to dryness (1,000× g, 10 min). Fresh water was used to rinse the polyamide several times till the water was nearly colorless. Then 95% ethanol was used to rinse the polyamide and the solution was collected. The ethanol solution was evaporated to dryness under reduced pressure to afford a solid residue (30 g (Table 2) spectral data are consistent with published data [29,30].  (Table 2) spectral data are consistent with published data [31,32].  (Table 2) spectral data are consistent with published data [32,33].  In this method, a microplate reader and 96 well plate were used to carry out the determination of the spectral absorption values. This assay is based on the classic method developed by Blois in 1958 [19]. Various forms of this method are widely used [36,37]. Unlike the commonly used methods, which are labor and time-consuming and reagent and sample-wasting, this microplate assay method is much more rapid, sample-saving and environmentally-friendly. In this method, methanolic DPPH solutions (100 μg/mL, 50 μL) were added to samples of different concentration (200 μL, 12.5-100 μg/mL).
These solutions were gently mixed and incubated in the dark for 30 min at room temperature. Then the absorbances of the resulting solutions were measured at 517 nm. For preparation of the standard curve, different concentrations of DPPH methanol solutions (5-50 μg/mL) were used. The DPPH concentration (μg/mL) in the reaction medium was calculated from the following calibration curve, determined by linear regression (r 2 : 0.9985): Absorbance (λ 517 ) = 0.022 × [DPPH · ] + 0.0145 The scavenging capability of test compounds was calculated using the following equation: where λ 517-C is absorbance of a control with no radical scavenger and λ 517-S is absorbance of the remaining DPPH in the presence of scavenger.

ABTS Assay
The ABTS assay was carried out using a method based on the original and classic method developed by Miller in 1993 [38] with some modifications. The ABTS radical should be preformed by reacting equal volumes of 1.1 mg/mL aqueous ABTS and 0.68 mg/mL potassium persulfate (K 2 S 2 O 8 ), and then storing in the dark for 6 h at room temperature, as described by Gülçin [6]. Then ABTS ·+ solutions (50 μL) were added to samples of different concentrations (200 μL, 12.5-100 μg/mL). These solutions were gently mixed and incubated in the dark for 30 min at room temperature. Then the absorbances of the resulting solutions were measured at 734 nm. Different concentrations of ABTS radical solutions (55-220 μg/mL) were used to prepare the standard curve. The ABTS radical concentration (μg/mL) in the reaction medium was calculated from the following calibration curve, determined by linear regression (r 2 : 0.9985): Absorbance (λ 734 ) = 0.0118 × [ABTS ·+ ] + 0.109 The scavenging capability of test compounds was calculated using the following equation: ABTS ·+ scavenging activity (%) = (1 − λ 734-S λ 734-C ) × 100 where λ 734-C is absorbance of a control with no radical scavenger and λ 734-S is absorbance of the remaining ABTS in the presence of scavenger.

Conclusions
Phenolic components of L. muscari were studied for the first time. Three amides, one flavone and one homoisoflavonone were isolated and their antioxidant activities were evaluated using two microplate assay methods. N-trans-feruloyltyramine (2) and N-trans-feruloyloctopamine (3) showed effective activity and the structure-activity relation investigation indicates that the -OCH 3 group at C-3 affects the antioxidant activity.