Protective Effect of Alpinia oxyphylla Fruit against tert-Butyl Hydroperoxide-Induced Toxicity in HepG2 Cells via Nrf2 Activation and Free Radical Scavenging and Its Active Molecules

Alpinia oxyphylla Miq. (Zingiberaceae) extract exerts protective activity against tert-butyl hydroperoxide-induced toxicity in HepG2 cells, and the antioxidant response element (ARE) luciferase activity increased 6-fold at 30 μg/mL in HepG2 cells transiently transfected with ARE-luciferase. To identify active molecules, activity-guided isolation of the crude extract led to four sesquiterpenes (1, 2, 5, 6) and two diarylheptanoids (3 and 4) from an n-hexane extract and six sesquiterpenes (7–12) from an ethyl acetate extract. Chemical structures were elucidated by one-dimensional, two-dimensional nuclear magnetic resonance (1D-, 2D-NMR), and mass (MS) spectral data. Among the isolated compounds, eudesma-3,11-dien-2-one (2) promoted the nuclear accumulation of nuclear factor (erythroid-derived 2)-like 2 (Nrf2) and increased the promoter property of the ARE. Diarylheptanoids, yakuchinone A (3), and 5′-hydroxyl-yakuchinone A (4) showed radical scavenging activity in 2,2-diphenyl-1-picrylhydrazyl (DPPH) and 3-ethylbenzothiazoline-6-sulphonic acid (ABTS) assays. Furthermore, optimization of extraction solvents (ratios of water and ethanol) was performed by comparison of contents of active compounds, ARE-inducing activity, radical scavenging activity, and HepG2 cell protective activity. As a result, 75% ethanol was the best solvent for the extraction of A. oxyphylla fruit. This study demonstrated that A. oxyphylla exerted antioxidant effects via the Nrf2/HO-1 (heme oxygenase-1) pathway and radical scavenging along with active markers eudesma-3,11-dien-2-one (2) and yakuchinone A (3).


Introduction
Excessive production of reactive oxygen species (ROS) may cause the degeneration of DNA, protein, and lipids that may be a contributing factor in the development of chronic diseases, including cancer, diabetes, atherosclerosis, and arthritis [1,2]. Strategies for protecting against oxidative damage are (i) scavenging ROS using antioxidants including polyphenols, carotenoids, and vitamins, (ii) up-regulation of endogenous antioxidant/phase II detoxifying enzymes through the induction of antioxidant signaling pathways such as the NF-E2-related factor-2 (Nrf2)/antioxidant response element (ARE) signaling pathway [3].

Plant Material
A. oxyphylla fruit was purchased from a Kyungdong Oriental Herb Market, Seoul, Korea, in June 2017 and identified by one of the authors (C.Y. Kim). A voucher specimen was deposited in the Herbarium of the College of Pharmacy, Hanyang University (HYU-AO-2017-06-01).

Extraction and Isolation of Compounds
A. oxyphylla fruit (1.2 kg) were ground into powder and extracted three times with methanol (3 L) for 90 min using an ultrasonic apparatus and subsequently filtered in vacuo through a defatted cotton filter. The filtered solute was then concentrated to dryness (126 g) by rotary evaporation under reduced pressure at 40 • C. Next, a crude methanol extract was suspended in 1000 mL water and successively partitioned with n-hexane, ethyl acetate (EtOAc), and n-butanol (1000 mL × 3 times) to obtain 36.5 g of n-hexane extract, 33.9 g of EtOAc extract, 12.5 g of n-butanol and 43.1 g of water extract, which was then stored in a freezer (−20 • C). The n-hexane extract was isolated by centrifugal partition chromatography (CPC) with a two-phase solvent system of n-hexane-EtOAcn-butanol-water (7:3:7:3, v/v) (Table S1). Firstly, the rotor was entirely filled, with the lower phase as the stationary phase. Subsequently, the upper phase was pumped into the inlet of the rotor at a flow rate of 10 mL/min while the apparatus ran at 1000 rpm. After reaching hydrostatic equilibrium, samples (10 g) were subjected to the CPC apparatus. The effluents were monitored at 254 nm, and nine fractions (Frs. A-I) were obtained according to the chromatographic profile. The CPC method was operated three times, and the same fractions were combined. For further purification of the Frs. C, D, H, and I, preparative high-performance liquid chromatography (prep-HPLC) was performed with an Inno C18 column (20 × 250 mm, 5 µm, YoungJin Biochrom, Seongnam, Korea) under isocratic conditions with 60% acetonitrile. After prep-HPLC, compounds 1-6 were obtained with high purities: compound 1 (181.6 mg) from Fr. C; compound 2 (6.2 mg) from Fr. D; compounds 3 (488.5 mg) and 4 (7.1 mg) from Fr. H. Compounds 5 (121.6 mg) and 6 (40.7 mg) were isolated from Fr. H after applying a silica gel column with n-hexane-ethyl acetate (7:1) Antioxidants 2022, 11, 1032 3 of 15 and subsequent prep-HPLC under isocratic 60% acetonitrile elution. A detailed isolation scheme is described in Figure S1.

Cell Culture
HepG2 cells were purchased from ATCC (Manassas, VA, USA) and cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 100 U/mL penicillin, and 100 µg/mL streptomycin in a humidified atmosphere of 5% CO 2 at 37 • C.

ARE-Luciferase Assay
HepG2 cells stably transfected with an ARE-driven reporter gene construct, pGL4.37[luc2P/ARE/Hygro], were kindly donated by Dr. IJ Cho (Daegu Haany University, Gyeongsan, Korea), who established the stable transfection of HepG2 cells with pGL4.37[luc2P/ARE/Hygro] construct as previously described [15]. HepG2 cells stably transfected with pGL4.37 were plated at a density of 1 × 10 5 cells/well in 24-well plates. After 24 h, cells were serum-starved overnight and subsequently treated with vehicle, extracts, and compounds for 12 h. Crude extract and compounds were dissolved in DMSO and added to the medium. The final concentration of DMSO did not exceed 0.1%, which did not affect cell cytotoxicity. Sulforaphane (5 µM) (Calbiochem, Darmstadt, Germany) was used as a positive control. After treatment, cells were washed twice with ice-cold phosphatebuffered saline and lysed in a passive lysis buffer (Promega, Madison, WI, USA). The levels of ARE-luciferase activity in the resultant cell lysates were measured using the luciferase assay system (Promega, Madison, WI, USA) according to the manufacturer's instructions.

Cytoprotective Effects of t-BHP-Induced HepG2 Cells
HepG2 cells were seeded in 96-well plates at a density of 7 × 10 4 cells/well. After 24 h, the culture medium (200 µL/well) was replaced with a serum-free medium (200 µL/well) containing various concentrations of extract (3, 10, and 30 µg/mL) or compounds 2-4 (3, 10, and 30 µM). After 12 h, the cells were treated with 300 µM t-BHP to induce oxidative stress for 3 h. The protective effect of extract or compounds was evaluated using an MTT assay. All experiments were performed at least in triplicate.

Preparation of Cell Lysates and Western Blotting
HepG2 cells were plated in 6-well plates at a density of 5 × 10 5 cells/well. After 24 h, the cells were incubated with different concentrations of extract or compounds for the Antioxidants 2022, 11, 1032 4 of 15 indicated time after overnight serum starvation. To obtain the nuclear and cytosolic cell lysates and detect heme oxygenase-1 (HO-1) and Nrf2, cells were incubated with 80 µL cold lysis buffer A, containing 10 mM-HEPES-KOH (pH 7.9), 10 mM-KCl, 1.5 mM-MgCl 2 , 1 mM-dithiothreitol, 0.1% nonidet P-40, and inhibitor stock III (1:200) for 10 min on ice. Next, samples were centrifuged at 7200× g, 4 • C for 5 min, and the supernatant fractions (cytosolic cell extracts) were collected to determine protein concentration. Pellets were re-suspended in 40 µL cold lysis buffer B, composed of 10 mM HEPES-KOH (pH 7.9), 25% glycerol, 400 mM-KCl, 0.1 mM-EDTA, and inhibitor III (1:200), and then kept on ice for 60 min. Samples were centrifuged at 7200× g, 4 • C for 10 min, and the supernatant fractions (nuclear cell extracts) were collected. Both cell extracts were analyzed for protein concentration and stored at −80 • C until use for Western blotting. Equal amounts (20 µg) were separated by 7.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred onto nitrocellulose membranes (Millipore, Bedford, MA, USA). The membranes were washed with Tris-buffered saline (10 mM Tris-HCl, 150 mM NaCl, pH 7.5), followed by blocking with 5% (w/v) non-fat dried milk. The membranes were incubated overnight with antibodies against HO-1, Nrf2, Lamin A/C, and actin at 4 • C. The membranes were washed three times with TBST buffer and then exposed to secondary antibodies coupled to horseradish peroxidase for 2 h at room temperature. ECL chemiluminescence reagents (Pierce; Thermo Fisher Scientific, Inc., Waltham, MA, USA) were used to detect the protein bands. Densitometric analysis of the data obtained from at least three independent experiments was performed using Fusion Solo chemiluminescence and fluorescence imaging system (Vilber-Lourmat, Eberhardzell, Germany).

Extraction for Optimization of Ethanol/Water Solvent Ratios
Fifty grams of A. oxyphylla powder were extracted with 100 mL of the following solvents: 100% water, 25% aqueous ethanol, 50% aqueous ethanol, 75% aqueous ethanol, and 100% ethanol for 2 h under reflux. The extract was filtered and dried under a rotary evaporator and freeze-dried.

HPLC Analysis
Each extract was dissolved at a concentration of 10 mg/mL in methanol. Isolated compounds 1-12 were used to prepare standard solutions. Each compound was dissolved at 1000 µg/mL concentration in methanol, and serially diluted (1000, 500, 250, 125, 62.5, 6.25, 3.125, 1.5625 and 0.78125 µg/mL). Linear regression equations were calculated with y = ax ± b, where x was the concentration and y was the peak area of each compound. Linearity was established by the coefficient of equation (R 2 ). HPLC analysis was performed using an Agilent Technologies 1200 system equipped with an automatic injector, a column oven, and a DAD detector. An Inno C18 column (250 × 4.6 mm, 5 µm, YoungJin Biochrom, Seongnam, Korea) was used. The temperature was maintained at 40 • C, with an injection volume of 20 µL at a flow rate of 1 mL/min. The mobile phase was composed of acetonitrile containing 0.1% formic acid (A) and water containing 0.1% formic acid (B). The gradient elution conditions were as follows: initial 0

DPPH Radical Scavenging Activity
The radical scavenging activity of 2,2-diphenyl-1-picrylhydrazyl (DPPH) was performed according to the method from Shim and Lee [16] that was slightly modified. In brief, 5 µL samples (final concentration: 25, 12.5, 6.25, 3.125 µg/mL in ethanol) were mixed with 100 µL of 400 mM DPPH in 100% (v/v) ethanol and 95 µL ethanol. After 1 h incubation in the dark covered with aluminum foil, a decrease in absorbance was monitored at 515 nm. The control consisted of 5 µL ethanol. Ethanol was used as a solvent, and trolox (final concentration: 1000, 500, 250, 125, 62.5 µg/mL) was used as a positive control. All samples were analyzed in triplicate.

ABTS Radical Scavenging Activity
The radical scavenging activity of 2,2-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) was assessed using a modified method [17]. In brief, ABTS (7 mM) and potassium persulfate (2.45 mM) were mixed and heated to 68 • C for 30 min with aluminum foil. After heating, the stock solution was diluted to 1/50 using distilled water. The reactants, 195 µL of the diluted ABTS radical solution and 5 µL of the sample (final concentration: 25, 12.5, 6.25, 3.125 µg/mL in ethanol), were mixed in 96-well plates. Absorbance was immediately read at 723 nm using a microplate reader. Ethanol was used as a solvent, and trolox (final concentration: 1000, 500, 250, 125, and 62.5 µg/mL) was used as a positive control. All samples were analyzed in triplicate.

Statistical Analysis
All data are reported as means ± S.D. The statistical significance of differences between treatments was assessed using Student's t-test. Probability values less than * < 0.05, ** < 0.01 were considered significant.
Compound 4 was obtained as a yellow gum. Its molecular formula was determined as C20H24O4 based on the ESI-MS ion peak at m/z 327.    Next, it was confirmed which compounds contributed to the increase in AR erase activity. The cytotoxic effects of compounds 1-12 were evaluated in HepG2 c compounds showed any cytotoxicity when treated at concentrations of 3-30 μM As shown in Figure 4, compounds 2, 3, and 4 from the n-hexane extract enhanced t luciferase activities 22.6-fold, 2.9-fold, and 4.4-fold at 30 μM, respectively. Sulfo (5 μM) induced ARE enhanced activity 42-fold. Since compounds 7, 9, and 12 from extract only slightly enhanced ARE-luciferase activity. Further studies were pe only using crude extract and compounds 2-4. Next, it was confirmed which compounds contributed to the increase in ARE-luciferase activity. The cytotoxic effects of compounds 1-12 were evaluated in HepG2 cells. No compounds showed any cytotoxicity when treated at concentrations of 3-30 µM for 24 h. As shown in Figure 4, compounds 2, 3, and 4 from the n-hexane extract enhanced the AREluciferase activities 22.6-fold, 2.9-fold, and 4.4-fold at 30 µM, respectively. Sulforaphane (5 µM) induced ARE enhanced activity 42-fold. Since compounds 7, 9, and 12 from EtOAc extract only slightly enhanced ARE-luciferase activity. Further studies were performed only using crude extract and compounds 2-4.

Effects of Crude Extract and Active Compounds 2-4 against t-BHP-Induced Cell Death
The protective effects of crude extract and compounds 2-4 were tested against t-BHP-induced toxicity of HepG2 cells. t-BHP converted free radicals are often used to cause oxidative stress [6,7]. In this study, 300 µM of t-BHP showed cytotoxicity and reduced cell viability to~20% of control levels, but pretreatment with crude extract or active compounds 2-4 for 12 h significantly prevented oxidative stress-induced cell death concentration-dependently ( Figure 5). Crude extract increased cell viability to 63.4% at 30 µg/mL concentration. Its active compounds also protect HepG2 cells against t-BHP treatment. Compound 2 at 100 µM concentration increases cell viability from 26.3% to 79.9%. Compounds 3 and 4, at a concentration of 30 µM, also improved the cell viability to 89.0% and 94.0%, respectively. In the induction of ARE-luciferase, compound 2 showed the most activity, but the protective ability of HepG2 cells against oxidative stress was better with compounds 3 and 4. Therefore, free radical scavenging activities were compared by DPPH and ABTS assays.

Effects of Crude Extract and Active Compounds 2-4 against t-BHP-Induced Cell Death
The protective effects of crude extract and compounds 2-4 were tested against t-BHPinduced toxicity of HepG2 cells. t-BHP converted free radicals are often used to cause oxidative stress [6,7]. In this study, 300 μM of t-BHP showed cytotoxicity and reduced cell viability to ~20% of control levels, but pretreatment with crude extract or active compounds 2-4 for 12 h significantly prevented oxidative stress-induced cell death concentration-dependently ( Figure 5). Crude extract increased cell viability to 63.4% at 30 μg/mL concentration. Its active compounds also protect HepG2 cells against t-BHP treatment. Compound 2 at 100 μM concentration increases cell viability from 26.3% to 79.9%. Compounds 3 and 4, at a concentration of 30 μM, also improved the cell viability to 89.0% and 94.0%, respectively. In the induction of ARE-luciferase, compound 2 showed the most activity, but the protective ability of HepG2 cells against oxidative stress was better with compounds 3 and 4. Therefore, free radical scavenging activities were compared by DPPH and ABTS assays.

Free Radical Scavenging Assay of Compounds (2-4)
ARE-inducing compounds 2-4 from n-hexane extract were evaluated by DPPH and ABTS radical scavenging activities at a concentration range of 3-100 μM. As shown in Figure 6, compound 3 exhibited the highest ABTS and DPPH radical scavenging activity,

Free Radical Scavenging Assay of Compounds (2-4)
ARE-inducing compounds 2-4 from n-hexane extract were evaluated by DPPH and ABTS radical scavenging activities at a concentration range of 3-100 µM. As shown in Figure 6, compound 3 exhibited the highest ABTS and DPPH radical scavenging activity, and compound 4 was next. However, compound 2 showed little DPPH and ABTS radical scavenging ability. Compared with induction of ARE-luciferase and free radical scavenging activities, compound 2 was considered to be an Nrf2 activator without radical scavenging ability, and compounds 3 and 4 behaved as free radical scavengers rather than ARE induction.

Nuclear Translocation of Nrf2 by Crude Extract and Compound 2
Compound 2 exhibited a protective effect against t-BHP-induced oxidative stress in HepG2 cells but did not scavenge free radicals in ABTS and DPPH assays. The compound 2 enhanced ARE-driven luciferase activity was described earlier. The nuclear translocation of Nrf2 was determined to confirm its ability to activate Nrf2. Treatment of compound 2 (30 μM) caused a 2-fold nuclear Nrf2 accumulation after 12 h exposure at a concentration of 30 μM ( Figure 7A). The treatment of compound 2 at 30 μM increased nuclear Nrf2 levels 2.3-fold, and 5 μM sulforaphane enhanced levels 3.9-fold. Then 30 μM of compound 2 was treated, and as a result of checking the change in the level of Nrf2 protein expression with time, it was confirmed that the increase was the highest at 3 h after increasing from 1 h, and it was maintained until 12 h. Furthermore, the representative Nrf2-controlled antioxidant protein, HO-1, was also increased by compound 2 and crude extract ( Figure 7C,D).

Nuclear Translocation of Nrf2 by Crude Extract and Compound 2
Compound 2 exhibited a protective effect against t-BHP-induced oxidative stress in HepG2 cells but did not scavenge free radicals in ABTS and DPPH assays. The compound 2 enhanced ARE-driven luciferase activity was described earlier. The nuclear translocation of Nrf2 was determined to confirm its ability to activate Nrf2. Treatment of compound 2 (30 µM) caused a 2-fold nuclear Nrf2 accumulation after 12 h exposure at a concentration of 30 µM ( Figure 7A). The treatment of compound 2 at 30 µM increased nuclear Nrf2 levels 2.3-fold, and 5 µM sulforaphane enhanced levels 3.9-fold. Then 30 µM of compound 2 was treated, and as a result of checking the change in the level of Nrf2 protein expression with time, it was confirmed that the increase was the highest at 3 h after increasing from 1 h, and it was maintained until 12 h. Furthermore, the representative Nrf2-controlled antioxidant protein, HO-1, was also increased by compound 2 and crude extract ( Figure 7C,D).

Optimization of Ethanol-Water Ratio for Extraction
Finally, the ethanol-water composition for extraction was optimized. The mixtures of ethanol and water were selected since these two solvents are preferable solvents due to safety considerations with health and handling.
We investigated the ethanol composition on extract characteristics such as radical scavenging (DPPH and ABTS) activity, ARE-induction effect, and HepG2 cell protective activity against t-BHP toxicity. In DPPH assays, 75% ethanol extract was the best, followed by 100% ethanol extract and 50% ethanol extract ( Figure 8A). In the ABTS assay, 75% ethanol extract showed the highest efficacy. Both the 25% ethanol extract and 100% ethanol extract also exhibited radical scavenging activity, but the efficacy was different depending on the concentration ( Figure 8B). When ARE-luciferase induction was tested, it was effective in the following order: 100% ethanol, 75% ethanol, and 50% ethanol ( Figure 8C).
As a result, the 75% ethanol extract most effectively prevented the toxicity of HepG2 cells induced by t-BHP ( Figure 8D). Therefore, 75% ethanol was selected as an extraction solvent for A. oxyphylla fruit.

Optimization of Ethanol-Water Ratio for Extraction
Finally, the ethanol-water composition for extraction was optimized. The mixtures of ethanol and water were selected since these two solvents are preferable solvents due to safety considerations with health and handling.
We investigated the ethanol composition on extract characteristics such as radical scavenging (DPPH and ABTS) activity, ARE-induction effect, and HepG2 cell protective activity against t-BHP toxicity. In DPPH assays, 75% ethanol extract was the best, followed by 100% ethanol extract and 50% ethanol extract ( Figure 8A). In the ABTS assay, 75% ethanol extract showed the highest efficacy. Both the 25% ethanol extract and 100% ethanol extract also exhibited radical scavenging activity, but the efficacy was different depending on the concentration ( Figure 8B). When ARE-luciferase induction was tested, it was effective in the following order: 100% ethanol, 75% ethanol, and 50% ethanol ( Figure 8C).
As a result, the 75% ethanol extract most effectively prevented the toxicity of HepG2 cells induced by t-BHP ( Figure 8D). Therefore, 75% ethanol was selected as an extraction solvent for A. oxyphylla fruit.
After comparing antioxidant activity, each solvent extract was analyzed by HPLC to assess the amount of isolated compounds 1-12. HPLC analysis revealed that nootkatone (1) and yakuchinone A (3) were major constituents in A. oxyphylla. Amount of eudesma-3,11dien-2-one (2), active molecule for ARE induction was 4.36 (mg/g extract) in 75% ethanol extract and 3.96 (mg/g extract) in 100% ethanol extract. Yakuchinone A (3) was extracted from 75% and 100% ethanol. However, in the remaining solvent conditions, it was below the analysis range. Since eudesma-3,11-dien-2-one (2) and yakuchinone A (3) were active molecules against t-BHP-induced toxicity in HepG2 cells, these amounts were carefully checked. These two compounds were extracted from 100% ethanol and 75% ethanol ( Table 1). 5 -Hydroxy-yakuchinone A (4) was also active, but the amount in all extracts was under the detectable range. Therefore, HPLC analysis made clear that eudesma-3,11-dien-2-one (2) and yakuchinone (3) were active markers in A. oxyphylla extract. After comparing antioxidant activity, each solvent extract was analyzed by HPLC to assess the amount of isolated compounds 1-12. HPLC analysis revealed that nootkatone (1) and yakuchinone A (3) were major constituents in A. oxyphylla. Amount of eudesma-3,11-dien-2-one (2), active molecule for ARE induction was 4.36 (mg/g extract) in 75% ethanol extract and 3.96 (mg/g extract) in 100% ethanol extract. Yakuchinone A (3) was extracted from 75% and 100% ethanol. However, in the remaining solvent conditions, it was below the analysis range. Since eudesma-3,11-dien-2-one (2) and yakuchinone A (3) were active molecules against t-BHP-induced toxicity in HepG2 cells, these amounts were carefully checked. These two compounds were extracted from 100% ethanol and 75% ethanol ( Table 1). 5′-Hydroxy-yakuchinone A (4) was also active, but the amount in all extracts was under the detectable range. Therefore, HPLC analysis made clear that eudesma-3,11dien-2-one (2) and yakuchinone (3) were active markers in A. oxyphylla extract.  A. oxyphylla fruit has been reported to exhibit antioxidant [27,28] and hepatoprotective effects against CCl 4 -induced oxidative damage via the Nrf2 pathway [29]. In addition, supercritical fluid extract of A. oxyphylla exhibited a strong ferric reducing power at the high dosage of 1000 µg/mL, but the extract showed a minor level of Fe 2+ scavenging effect [30]. 5-Hydroxymethylfurfural [30], protocatechuic acid [31], diarylheptanoids including yakuchinone A [32], and flavonoids [33] were reported as antioxidant active substances. Yakuchinone A was reported to possess DPPH radical scavenging and linoleic acid peroxidation inhibitory activities [34]. Nootkatone, the principal ingredient, was also reported to activate Nrf2 in the luciferase assay using transfected HCT-116 cells and increased HO-1 protein levels as a marker for antioxidant activity [28].

Conclusions
A. oxyphylla extract significantly protects HepG2 cells against t-BHP-induced toxicity in a concentration-dependent manner. Bioassay-guided isolation led to the identification of two main active components, eudsma-3,11-dien-2-one (2) and yakuchinone A (3). Edesma-3,11-dien-2-one (2) promoted the nuclear accumulation of Nrf2 and increased the HO-l expression, and yakuchinone A (3) scavenged ABTS and DPPH radicals. Extraction solvent with mixtures of ethanol and water was also tested. Considering HepG2 cells protection, the ability of radical scavenging, ARE induction activities, and contents of active compounds 2, 3, 75% ethanol was chosen as the best extraction solvent. These results demonstrated that A. oxyphylla exerted antioxidant effects via the Nrf2/ARE pathway and radical scavenging, and the active markers were eudesma-3,11-dien-2-one (2) and yakuchinone A (3). Therefore, A. oxyphylla has the potential for use as an effective natural antioxidant for the prevention and treatment of oxidative stress-related liver diseases.
Supplementary Materials: The following supporting information can be downloaded at: www.mdpi.com/article/10.3390/antiox11051032/s1, Figure S1: Isolation scheme of n-hexane extract of A. oxyphylla; Figure S2: Isolation scheme of ethyl acetate extract; Figure S3

Conclusions
A. oxyphylla extract significantly protects HepG2 cells against t-BHP-induced toxicity in a concentration-dependent manner. Bioassay-guided isolation led to the identification of two main active components, eudsma-3,11-dien-2-one (2) and yakuchinone A (3). Edesma-3,11-dien-2-one (2) promoted the nuclear accumulation of Nrf2 and increased the HOl expression, and yakuchinone A (3) scavenged ABTS and DPPH radicals. Extraction solvent with mixtures of ethanol and water was also tested. Considering HepG2 cells protection, the ability of radical scavenging, ARE induction activities, and contents of active compounds 2, 3, 75% ethanol was chosen as the best extraction solvent. These results demonstrated that A. oxyphylla exerted antioxidant effects via the Nrf2/ARE pathway and radical scavenging, and the active markers were eudesma-3,11-dien-2-one (2) and yakuchinone A (3). Therefore, A. oxyphylla has the potential for use as an effective natural antioxidant for the prevention and treatment of oxidative stress-related liver diseases.