Flavonoids from Machilus japonica Stems and Their Inhibitory Effects on LDL Oxidation

Stems of Machilus japonica were extracted with 80% aqueous methanol (MeOH) and the concentrated extract was successively extracted with ethyl acetate (EtOAc), normal butanol (n-BuOH), and water. Six flavonoids were isolated from the EtOAc fraction: (+)-taxifolin, afzelin, (−)-epicatechin, 5,3'-di-O-methyl-(−)-epicatechin, 5,7,3'-tri-O-methyl-(−)-epicatechin, and 5,7-di-O-methyl-3',4'-methylenedioxyflavan-3-ol. The chemical structures were identified using spectroscopic data including NMR, mass spectrometry and infrared spectroscopy. This is the first report of isolation of these six compounds from M. japonica. The compounds were evaluated for their diphenyl picryl hydrazinyl scavenging activity and inhibitory effects on low-density lipoprotein oxidation. Compounds 1 and 3–6 exhibited DPPH antioxidant activity equivalent with that of ascorbic acid, with half maximal inhibitory concentration (IC50) values of 0.16, 0.21, 0.17, 0.15 and 0.07 mM, respectively. The activity of compound 1 was similar to the positive control butylated hydroxytoluene, which had an IC50 value of 1.9 µM, while compounds 3 and 5 showed little activity. Compounds 1, 3, and 5 exhibited LDL antioxidant activity with IC50 values of 2.8, 7.1, and 4.6 µM, respectively.


Introduction
The family Lauraceae includes about 32 genera and 2500 species that are distributed in tropical and subtropical regions, especially in Southeast Asia [1]. In Korea, members of the Lauraceae family are found in the southern regions and in Jeju Island. In this study, we describe the chemical constituents and activity of Machilus japonica extracts. M. japonica is an evergreen tree with yellow-green flowers and oval leaves. A few studies on the chemical composition of this plant have resulted in the identification of lignans [2] and terpene compounds [3]. In addition, the biological activities of M. japonica extracts such as insecticidal activity [4], elastase inhibition activity [5], and inhibition of matrix metalloproteinase-9 activity [6] have been investigated; however, all of these activities have been identified from extracts of M. japonica leaves. Taking a different approach, the aim of this study was to explore the active components of the stem of M. japonica and to screen their biological activities.
A total of six flavonoids were isolated and identified from the stem of M. japonica. Because many flavonoids have been reported to exhibit antioxidant activities [7], the isolated flavonoids were also evaluated for DPPH radical scavenging activity. low-density lipoprotein (LDL) is susceptible to oxidative damage, and oxidized LDL (oxLDL) plays a key role in the development of atherosclerotic lesions [8]. oxLDL in vessel walls is subjected to rapid uptake by scavenger receptors on monocyte derived macrophages, leading to the formation of foam cells that accumulate cholesterol [9]. Thus, we also evaluated the isolated compounds for their ability to inhibit LDL oxidation. Our results suggest that extracts of M. japonica stems and the specific flavonoids found in these extracts may prove useful for preventing or treating hypercholesterolemia and atherosclerosis.
Compound 5 was obtained as colorless needles and a molecular ion peak [M] + was observed at m/z 332 in the EI/MS spectrum. The IR spectrum showed absorbance bands for hydroxyl (3412 cm −1 ) and aromatic (1616 cm −1 ) groups. The 1 H-NMR and 13 C-NMR spectra of compounds 4 and 5 were analogous, showing (−)-epicatechin skeleton signals. Due to the m/z of 332 and observation of three methoxy groups, compound 5 was presumed to be (−)-epicatechin-tri-O-methylate. To confirm the position of the methoxy groups, a gHMBC experiment was conducted. Judging by the correlation between the three methoxy proton signals at δH 3.69 (6H, s) and 3.66 (3H, s) with three oxygenated olefin quaternary carbon signals at δC 160.0 (C-7), δC 159.6 (C-5), and δC 148.3 (C-3') in the gHMBC spectrum, compound 5 was identified as 5,7,3'-tri-O-methyl-(−)-epicatechin, which was confirmed by comparison with spectroscopic data in the literature [15].

Evaluation of Radical Scavenging Activity and Inhibitory Effect on LDL Oxidation
Compounds 1 and 3-6 exhibited DPPH radical scavenging with IC50 values of 0.16, 0.21, 0.17, 0.15, and 0.07 mM, respectively, which were equivalent with that of ascorbic acid (0.18 mM) ( Table 1). The lack of scavenging activity by compound 2 was speculated to be due to the glucosyl moiety on the C ring [17]. Compound 6 exhibited the highest antioxidant activity, which we attributed to the dioxymethylene moiety on the B ring. Highly reactive molecules called free radicals can cause tissue damage by reacting with polyunsaturated fatty acids in cellular membranes, nucleotides in DNA, and critical sulfhydryl bonds in proteins. In addition to cellular damage, cataract formation, photodermatoses, aging, and inflammatory diseases such as arthritis are associated with free radicals [18]. Therefore, the search for compounds from natural sources that can protect against free radicals by endogenous and exogenous antioxidants is of special significance for human health.
Antioxidants act by donating hydrogen atoms to lipid radicals. Radicals obtained from antioxidants with molecular structures such as phenols are stable species that can halt the oxidation chain reaction [19]. To determine whether these compounds might be effective in the development of hypercholesterolemic or antiatherogenic agents, their potential for inhibiting LDL oxidation was evaluated.
Compounds 1, 3, and 5 demonstrated LDL antioxidant activity with IC50 values of 2.8, 7.1, and 4.6 µM, respectively (Table 2). Compound 1 was similar to the positive control, BHT, which had an IC50 value of 1.9 µM, while compounds 3 and 5 also showed significant activity. Compound 4 exhibited a low level of LDL antioxidant activity with an IC50 value of 79.1 µM. Compounds 2 and 6 were not effective as LDL antioxidants, which was attributed to the glucosyl moiety at C-3 for compound 2 and steric hindrance of the hydroxyl groups at C-3' and C-4' by dioxymethylene moieties in compound 6. Table 2. Inhibitory effects of flavonoids isolated from the stems of Machilus japonica on low-density lipoprotein (LDL) oxidation. a The IC50 value of each compound was defined as the concentration (μM) that resulted in 50% inhibition of LDL oxidation. The results are averages of three independent experiments, and the data are expressed as the mean ± SD. LDL oxidation is regarded as a key step in the formation of atherosclerotic lesions [20,21]. Experimental evidence demonstrating an association between oxLDL cholesterol and both the presence of atherosclerotic lesions and progression of carotid artery atherosclerosis support this hypothesis [22,23]. Vitamin E, one of the most popular natural antioxidants, inhibits atherogenesis by inhibiting LDL oxidation (IC50: 2.4 μM) [24]. Several other effective natural dietary antioxidants comprising phenolic compounds and carotenoids have been identified as well, in addition to vitamins and enzymes [25]. The inhibitory activity of compounds 1, 3, and 5 was very similar to that of the well-known antioxidant BHT (IC50: 1.9 μM), and showed significant antioxidant capacity relative to several other naturally-occurring antioxidants such as nectandrin B (IC50: 4.1 μM) from tabu (Machilus thunbergii) [26] and (+)-lariciresinol (IC50: 11.9 μM) from Rousa dogwood (Cornus kousa) [27]. Therefore, the flavonoids isolated from Machilus japonica may be a good natural source of antiatherogenic agents.

Plant Materials
Dried stems of Machilus japonica were supplied by GFC Co., Ltd., Suwon, Korea in January 2010, and were identified by Dae-Keun Kim, College of Pharmacy, Woosuk University, Jeonju, Korea. A voucher specimen (KHU2010-0103) has been reserved at the Laboratory of Natural Product Chemistry, Kyung Hee University, Yongin, Korea.

General Experimental Procedures
Melting points were determined using a Fisher-John's Melting Point apparatus (Fisher Scientific, Miami, FL, USA) with a microscope and the values obtained were uncorrected. Optical rotations were measured using a JASCO P-1010 digital polarimeter (Tokyo, Japan). NMR spectra were recorded on a 400 MHz FT-NMR spectrometer (Varian Inova AS 400, Palo Alto, CA, USA). IR spectra were obtained from a Perkin Elmer Spectrum One FT-IR spectrometer (Buckinghamshire, England). FAB-MS data were recorded on a JEOL JMS-700 (Tokyo, Japan), EI-MS on a JEOL JMSAX 505-WA (Tokyo, Japan), and ESI-MS on a Finnigan LCQ Advantage spectrometer (Thermo Scientific, Waltham, MD, USA). The UV lamp used was a Spectroline Model ENF-240 C/F (Spectronics Corporation, Westbury, NY, USA). Kiesel gel 60 silica gel resin was used for column chromatography (c.c.) (Merck, Darmstadt, Germany) and the ODS was a LiChroprep RP-18 (Merck). TLC analysis was carried out using Kiesel gel 60 F254 and RP-18 F254S (Merck). Deuterated solvents were purchased from Merck Co. Ltd. and Sigma Aldrich Co. Ltd. (St. Louis, MO, USA).

DPPH Radical Scavenging Activity
The DPPH radical scavenging activity assay was based on the capacity of a substance to scavenge stable DPPH radicals. Briefly, reaction mixtures containing test samples (100 µL) and an ethanolic DPPH solution (100 µL of 0.06 mM) were placed in 96-well microplates and incubated at 37 °C for 30 min. Absorbance values were measured at 517 nm.

LDL Isolation and Oxidation Assay
Plasma was obtained from fasted healthy normalipidemic volunteers. LDL isolation and TBARS assays were performed as previously described with slight modification [28]. Briefly, an LDL solution (250 μL, 50-100 μg protein) in 10 mM PBS (pH 7.4) was supplemented with 10 μM CuSO4. Oxidation was performed in screw-capped 5-mL glass vials at 37 °C in the presence of either compound 1, 3, or 5. After incubation for 4 h the reaction was terminated by the addition of 1 mL 20% TCA. Following precipitation, 1 mL 0.67% TBA in 0.05 N NaOH was added and the mixture was vortexed, after which the final mixture was heated for 5 min at 95 °C, cooled on ice, and centrifuged for 2 min at 1000× g. The optical density of MDA generated in the assay was measured at 532 nm. Calibration was performed using an MDA standard prepared from tetramethoxypropane.