Xanthine Oxidase Inhibitory Triterpenoid and Phloroglucinol from

Dec 15, 2010 - Arctium minus crude extract presents antinociceptive effect in a mice acute gout attack model. Susana Paula Moreira Fischer , Indiara B...
2 downloads 0 Views 2MB Size
Download PDF
J. Agric. Food Chem. 2011, 59, 407–414

407

DOI:10.1021/jf1041382

Xanthine Oxidase Inhibitory Triterpenoid and Phloroglucinol from Guttiferaceous Plants Inhibit Growth and Induced Apoptosis in Human NTUB1 Cells through a ROS-Dependent Mechanism

†

)

)

KAI-WEI LIN,†,^ A-MEI HUANG,‡,^ HUANG-YAO TU,† LING-YI LEE,† CHIEN-CHANG WU,† TZYH-CHYUAN HOUR,‡ SHYH-CHYUN YANG,*,† YEONG-SHIAU PU,§ AND CHUN-NAN LIN*, School of Pharmacy, ‡Institute of Biochemistry, College of Medicine, and Faculty of Fragrance and Cosmetics, College of Pharmacy, Kaohsiung Medical University, Kaohsiung 807, Taiwan, and § Department of Urology, College of Medicine, National Taiwan University, Taipei 100, Taiwan. ^ These authors contributed equally to this work

A known triterpenoid, β-amyrin (1), and a known and a new phloroglucinol, cohulupone (2) and garcinielliptone P (3), were isolated from the pericarp and heartwood and seed of Garcinia subelliptica, respectively. A new xanthonolignoid, hyperielliptone HF (4), was isolated from the heartwood of Hypericum geminiflorum. The new compounds were established by analysis of their spectroscopic data. Compounds 1-3 showed an inhibitory effect on xanthine oxidase (XO). Treatment of NTUB1, a human bladder cancer cell, with 1 or 1 cotreated with cisplatin for 24 h resulted in a decreased viability of cells. Exposure of NTUB1 to 1 or 1 cotreated with cisplatin for 24 h significantly increased the level of production of reactive oxygen species (ROS). Flow cytometric analysis indicated that treatment of NTUB1 with 1 or 1 cotreated with cisplatin led to the cell cycle arrest, accompanied by an increase in the extent of apoptotic cell death in 1 or 1 combined with cisplatin-treated NTUB1 after 24 h. These data suggested that the presentation of cell cycle arrest and apoptosis in 1 or 1 combined with cisplatin-treated NTUB1 for 24 h was mediated through an increased amount of ROS in cells exposed to 1 or 1 cotreated with cisplatin. KEYWORDS: Garcinia subelliptica; Hypericum geminiflorum; xanthine oxidase; reactive oxygen species; cytotoxicity

INTRODUCTION

In previous papers, we isolated several new terpenoids, xanthonolignoids, and phloroglucinols from Guttiferaceous plant, Garcinia subelliptica Mett. and Hypericum geminiflorum Hemsl (1, 2), and reported their antiinflammatory activity, cytotoxicity, and DNA strand scission (1-5). Recently, we reported that phloroglucinols and synthesized triterpenoids showed significant antioxidant activity (5-7). It has been reported that antioxidants, such as pyrrolidine dithiocarbamate, epigallocatechin gallate, genistein, and vitamin E combined with a low dose of anticancer agents could enhance the cell death and reduce the side effect of anticancer agents (8). Cisplatin, a platinum complex, is an effective anticancer agent and is one of the most widely used drugs either alone or combination with other chemotherapeutic agents. However, cisplatin has several side effects including nephrotoxicity and cisplatin resistance in clinically used chemotherapeutic agents (9).

To study the cytotoxic effect and mechanism of induced cell death of natural antioxidants and the potential effectiveness of a combination of anticancer agents with natural antioxidants, the authors have further isolated a known triterpenoid, β-amyrin (1), a known phloroglucinol, cohulupone (2), and a new phloroglucinol, garcinielliptone P (3), and a new xanthonolignoid, hyperielliptone HF (4) (Figure 1) from G. subelliptica and H. geminiflorum, respectively. In this work, structure elucidation of 3 and 4, antioxidant activity of 1-4, and cytotoxic activity of 1 and 5/5a (Figure 1), previously reported as XO inhibitor (2), are reported. MATERIALS AND METHODS

*Corresponding authors. S.-C.Y.: e-mail, [email protected]. C.-N.L.: tel, þ886 7 3121101; fax, þ886 7 5562365; e-mail, lincna@ cc.kmu.edu.tw.

General Procedures. Optical rotations were recorded on a JASCO370 polarimeter using appropriate solvent. UV spectra were obtained in MeOH on a JASCO model 7800 UV-vis spectrophotometer. IR spectra were measured on a Hitachi 260-30 spectrophotometer. 1H (400 MHz) and 13 C NMR (100 MHz) spectra and 1H-1H COSY, NOESY, HMQC, and HMBC experiments were recorded on a Varian Unity-400 NMR spectrometer. MS were obtained on a JMS-HX100 mass spectrometer. Silica gel (Merck), particle size 15-40 μM, was used for column chromatography. Silica gel 60 F254 precoated aluminum sheets (0.2 mm, Merck) were employed for TLC. All solvents were of HPLC grade.

© 2010 American Chemical Society

Published on Web 12/15/2010

pubs.acs.org/JAFC

408

J. Agric. Food Chem., Vol. 59, No. 1, 2011

Lin et al.

Figure 2. Selective 2D NMR correlations for 4.

Figure 1. Structures of 1-5/5a. Chemicals. Xanthine, xanthine oxidase, allopurinol, 2,20 -azinobis(3-ethylbenzothiazoline-6-sulfonic acid) (ABTS), potassium persulfate, and tocopherol were purchased from Sigma Chemicals. Cisplatin was obtained from Pharmacia and Upjohn, Milan, Italy. All culture reagents were obtained from Gibco BRL. Plant Material. The heartwoods of G. subelliptica (10 kg) and H. geminiflorum (9.8 kg) and the fruit of G. subelliptica (22.8 kg) were collected at Ping-Tung Hsien, Taiwan, in August 2004 and October 2003 and at Kaohsiung, Taiwan, in July 2001 and authenticated by Dr. Ming-Hong Yen, School of Pharmacy, Kaohsiung Medical University, respectively. Voucher specimens (2004-GH) and (2003-5) and (2003) were deposited in the Laboratory of Medicinal Chemistry, School of Pharmacy, Kaohsiung Medical University. Extraction and Isolation. The fresh pericarp of G. subelliptica (15.3 kg) was extracted with CHCl3 (20 L) at room temperature. The CHCl3 extract was concentrated under reduced pressure to afford a brown residue (196 g). The residue was fractionated by chromatography over silica gel using cyclohexane-EtOAc (9: 1) to yield 1 (30 mg). The heartwood of G. subelliptica (10 kg) was chipped and extracted with CH2Cl2 (10 L) at room temperature. The resultant dried CH2Cl2 extract (100 g) was chromatographed over a silica gel column and eluted with n-hexane containing increasing amounts of EtOAc and with a final wash using MeOH to yield 33 fractions. Fraction 12 was further purified on a RP18 column eluted with MeOH to yield 2 (52 mg). The seeds (7.5 kg), obtained from the fresh fruits (22.8 kg) of G. subelliptica, were extracted with chloroform (10 L) at room temperature. The CHCl3 extract was dried under reduced pressure to afford a brown residue (130 g). This residue was fractionated by chromatography over silica gel and eluted with a gradient of n-hexane-EtOAcMeOH (4:4:1) to n-hexane-EtOAc-MeOH (1:2:1) to yield two fractions. Fraction 1 was subjected to repeated chromatography on silica gel and eluted with n-hexane-acetone (3:1) to yield three fractions. Fraction 3 was subjected to repeated chromatography on silica gel and eluted with CHCl3-EtOAc (9:2) to yield 3 (15 mg). The air-dried heartwood of

H. geminiflorum (9.8 kg) was extracted with CHCl3 (10 L). The CHCl3 extract was dried under reduced pressure to give a dark brown residue (40 g). This residue was fractionated by chromatography over silica gel and eluted with a gradient of n-hexane-EtOAc (4:1) to n-hexane-EtOAc (1:2) to yield four fractions. Fraction 2 was subjected to repeated chromatography on silica gel and eluted with n-hexane-EtOAc (3:1) to give 5/5a (56 mg) (5). Fraction 3 was subjected to repeated chromatography on silica gel and eluted with n-hexane-EtOAc (2:1) to yield 4 (7 mg). β-Amyrin (1). Colorless needles, mp 189-191 C; [R]25 D 57 (c 0.15, CHCl3); IR (KBr) νmax 3420 cm-1; 1H NMR (CDCl3) δ 0.79 (3H, s, Me-28), 0.83 (3H, s, Me-24), 0.87 (6H, s, Me-29, 30), 0.94 (3H, s, Me-25), 0.96 (3H, s, Me-26), 0.99 (3H, s, Me-23), 1.13 (3H, s, Me-27), 3.22 (1H, dd, J = 10.8, 4.4 Hz, H-3), 5.18 (1H, t, J = 3.6 Hz, H-12); 13C NMR (CDCl3) δ 15.5 (Me-24), 15.6 (Me-25), 16.8 (Me-26), 18.3 (C-6), 23.5 (C-11), 23.7 (Me-30), 26.1 (C-16), 26.9 (Me-27), 27.2 (C-2), 28.1 (Me-23), 28.4 (C-15), 29.7 (Me-28), 31.1 (C-20), 32.5 (C-17), 32.6 (C-7), 33.3 (Me-29), 34.7 (C-21), 36.9 (C-22), 37.1 (C-10), 38.6 (C-1), 38.8 (C-4), 39.8 (C-8), 41.7 (C-14), 46.8 (C-19), 47.2 (C-18), 47.6 (C-9), 55.1 (C-5), 78.9 (C-3), 121.7 (C-12), 145.2 (C-13); EIMS (70 eV) m/z 426 [M]þ (29), 411(1), 219 (18), 218 (100), 203 (44), 189 (17) (10). Cohulupone (2). White powder; UV (MeOH) λmax (log ε) 250 (3.74), 320 (3.63) nm; IR (KBr) νmax 3430, 1750, 1613 cm-1; 1H NMR (CD3OD) δ 1.03 (6H, d, J=6.8 Hz, Me-18 and Me-19), 1.51 (12H, s, Me-9, 10, 14, and 15), 2.29 (4H, m, H2-6 and 11), 3.80 (1H, m, H-17), 4.85 (2H, t, J=6.8 Hz, H-7 and 12); 13C NMR (CD3OD) δ 18.5 (Me  2), 19.5 (Me-18 and 19), 26.7 (Me  2), 34.3 (C-6 and 11), 38.4 (C-17), 56.3 (C-2), 119.2 (C-7 and 12), 122.6 (C-4), 136.0 (C-8 and C-13), 183.1 (C-5), 200.0 (C-3 and 16), 205.3 (C-1); ESIMS m/z 341 [M þ Na]þ, 319; HRESIMS m/z 341.1728 [M þ Na]þ (calcd for C19H26O4Na, 341.1729). Garcinielliptone P (3). Colorless oil; [R]25 D -2 (c 1.6, CHCl3); UV (MeOH) λmax (log ε) 280 (4.58) nm; IR (film on NaCl) νmax 3416 (OH), 1727 (CdO), 1667 (conjugated CO), 1620 (CdC) cm-1; 1H NMR (CDCl3) δ 1.00 (s, H3-10), 1.03 (d, J = 6.4 Hz, H3-30), 1.13 (d, J = 6.4 Hz, H3-29), 1.18 (m, HR-12), 1.23 (s, H3-11), 1.29 (s, H3-20), 1.31 (s, H3-21), 1.35 (m, H-8), 1.45 (d, J=13.2 Hz, HR-7), 1.53 (s, H3-16), 1.64 (s, H3-15), 1.65 (s, H3-25), 1.69 (s, H3-26), 1.85 (dd, J=13.2, 4.0 Hz, Hβ-7), 2.08 (m, Hβ-12), 2.19 (m, H-28), 2.47 (dd, J = 14.8, 7.6 Hz, HR-22), 2.55 (dd, J = 14.8, 7.6 Hz, Hβ-22), 4.43 (d, J = 4.4 Hz, Hβ-18), 4.89 (t, J = 7.2 Hz, H-13), 5.11 (t, J=7.2 Hz, H-23), 5.37 (d, J=4.4 Hz, Hβ-17); 13C NMR (CDCl3) δ 15.7 (C-9), 17.8 (C-16), 18.1 (C-26), 20.4 (C-29), 21.4 (C-30), 23.1 (C-11), 24.6 (C-21), 25.0 (C-20), 25.8 (C-15), 26.0 (C-25), 26.5 (C-12), 28.5 (C-22), 38.5 (C-7), 42.4 (C-28), 43.1 (C-8), 46.4 (C-9), 54.9 (C-2), 70.5 (C-17), 71.2 (C-19), 83.2 (C-6), 99.2 (C-18), 118.0 (C-23), 121.6 (C-4), 122.3 (C-13), 133.8 (C-14), 135.3 (C-24), 179.2 (C-3), 188.3 (C-5), 205.5 (C-1), 208.7 (C-27); EIMS m/z 500 [M]þ (5), 482 (9), 464 (10), 371 (35), 353 (30), 313 (100), 298 (60), 287 (47), 258 (49), 233 (44), 203 (51), 69 (83); HREIMS m/z 500.3137 [M]þ (calcd for C30H44O6, 500.3138). Hyperielliptone HF (4). Yellow powder; [R]25 D 0 (c 0.1, acetone); UV (MeOH) λmax (log ε) 212 (4.40), 245 (4.10), 265 (3.90), 317 (3.70) nm; IR (KBr) νmax 3443 (OH), 1633 (CdO) cm-1; 1H NMR (pyridine-d5) δ 3.81 (s, OMe-300 and OMe-500 ), 4.01 (dd, J = 12.8, 3.6 Hz, CHHOH), 4.08 (s, OMe-2), 4.37 (dd, J=12.8, 2.4 Hz, CHHOH), 4.57 (m, H-50 ), 5.75 (d, J= 8.0 Hz, H-60 ), 7.22 (s, H-200 and H-600 ), 7.35 (td, J=8.0, 1.2 Hz, H-6), 7.46 (d, J = 8.0 Hz, H-5), 7.70 (td, J = 8.0, 1.2 Hz, H-7), 8.40 (dd, J = 8.0, 1.2 Hz, H-8), 13.03 (s, OH-1); 13C NMR (pyridine-d5) 56.4 (C-12 and C-13), 61.0 (C-11), 61.2 (CH2OH), 78.8 (C-60 ), 79.3 (C-50 ), 103.9 (C-4a), 106.3

Article

Figure 3. Dose-dependent inhibition of XO by 1-3, 5/5a (2), and allopurinol. Data are presented as the means ( SD, n = 3-6. p < 0.05 (a) and p < 0.01 (b) compared to the control value, respectively.

J. Agric. Food Chem., Vol. 59, No. 1, 2011

409

Figure 6. Cisplatin, 5/5a, and cisplatin combined with 5/5a induced cell death. Cell viability was assessed by MTT assay for 72 h after treatment with 5 μM cisplatin, different concentrations of 5/5a, and 5 μM cisplatin combined with different concentrations of 5/5a. p < 0.05 (a) and p < 0.01 (b) compared to the control value, respectively.

Figure 4. ABTS radical scavenging activity of different concentrations of 4 and TOC. Values are means ( SD (n = 3). p < 0.05 (a), p < 0.01 (b), and p < 0.001 (c) compared to the control value, respectively.

Figure 7. Cisplatin and a combination of cisplatin and 1 induced cell death. Cell viability was assessed by MTT assay for 72 h after treatment with different concentrations of cisplatin and different concentrations of cisplatin combined with different concentrations of 1. p < 0.05 (a), p < 0.01 (b), and p < 0.001 (c) compared to the control value, respectively.

Figure 5. Compound 1 induced NTUB1 cell death. Cell viability was assessed by the MTT assay for 72 h after treatment with different concentrations of 1. p < 0.05 (a), p < 0.01 (b), and p < 0.001 (c) compared to the control value. (C-200 and C-600 ), 118.0 (C-5), 120.6 (C-8a), 124.5 (C-7), 126.1 (C-8), 126.5 (C-100 ), 130.8 (C-2), 131.0 (C-4), 131.7 (C-2), 135.2 (C-6), 138.6 (C-400 ), 141.3 (C-4a), 148.1 (C-1), 149.4 (C-300 and C-500 ), 156.2 (C-10a), 181.6 (C-9); ESIMS m/z 505 [M þ Na]þ, 453; HRESIMS m/z 505.1114 [M þ Na]þ (calcd for C25H22O10Na, 505.1111).

Hyperielliptone HB (5/5a). Colorless oil; [R]25 D 1.0 (c 1.0, CHCl3); UV (MeOH) λmax (log ε) 215 (4.2), 226 (4.1), 280 (4.3) nm; IR (film on NaCl) νmax 3417, 1714, 1673, 1589 cm-1; ESIMS m/z 483 ([M þ Na]þ), 437, 381, 288; HRESIMS m/z 483.2720 [M þ Na]þ (calcd for C27H40O6Na, 483.2722) (5). Assay of xanthine oxidase activity, free radical scavenging activity, and determination of ABTS radical cation scavenging capacity were performed by methods described previously (5, 11). Cell Culture and MTT Assay for Cell Viability. NTUB1, a human bladder cancer cell line, was maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum (FBS), 100 units/mL penicillin G, 100 μg/mL streptomycin, and 2 mM L-glutamine. The cells were cultured at 37 C in a humidified atmosphere containing 5% CO2. For evaluating the cytotoxic effect of 1, 1 combined with cisplatin, 5/5a, 5/5a combined with cisplatin, and positive control cisplatin, a modified 3-[4,5-dimethylthiazol2-yl]-2,5-diphenyltetrazolium bromide (MTT; Sigma Chemical Co.) assay was performed (12). Briefly, the cells were plated at a density of 1800 cells/ well in 96-well plates and incubated at 37 C overnight before drug exposure.

410

J. Agric. Food Chem., Vol. 59, No. 1, 2011

Lin et al.

Figure 8. Effect of 1 on the production of ROS in NTUB1 cells: control (A), 5 μM cisplatin (B), 10 μM cisplatin (C), 25 μM 1 (D), 50 μM 1 (E), and 75 μM 1 (F), for 24 h. The amount of ROS was assayed by H2DCFDA staining. Each sampling measured the fluorescence intensity of region M1 of 3  105 cells cotreated via autofluorescence. The control cells were treated with medium. Results were repeated by independent experiments. Cells were then cultured in the presence of graded concentrations of 1, 5/5a, 1 combined with 5 and 10 μM cisplatin (Pharmacia and Upjohn), 5/5a combined with 5 μM cisplatin, and 5 and 10 μM cisplatin at 37 C for 72 h. At the end of the culture period, 50 μL of MTT (2 mg/mL in PB) was added to each well and allowed to react for 3 h. Following centrifugation of plates at 1000g for 10 min, media were removed, and 150 μL of DMSO was added to each well. The proportions of surviving cells were determined by absorbance spectrometry at 540 nm using MRX (DYNEXCO) microplate reader. The cell viability was expressed as a percentage to the viable cells of control culture condition. The IC50 values of each group were calculated by the median effect analysis and presented as mean ( standard deviation (SD). Quantitative Analysis of Intracellular Reactive Oxygen Species (ROS). Production of ROS was analyzed by flow cytometry as described previously (13). Briefly, cells were plated and treated as indicated conditions. Ten micromolar 2,7-dichlorodihydrofluorescein diacetate (H2DCFHDA; Molecular Probes, Eugene, OR) was added to the treated cells 30 min prior harvest. The cells were collected by trypsinization and washed with PBS. The green fluorescence of intracellular DCF (20 ,70 -dichlorofluorescein) was then analyzed immediately by FACScan flow cytometer with a 525 nm band-pass filter (Becton Dickinson). Flow Cytometry Analysis. DNA content was determined following propidium iodide (PI) staining of cells as previously described (14). Briefly, 8  105 cells were plated and treated with 5 and 10 μM cisplatin, various concentrations of 1, and different concentrations of 1 combined with 5 or 10 μM cisplatin for 24 h, respectively. These cells were harvested by trypsinization, washed with 1  PBS, and fixed in ice-cold MeOH at

-20 C. After overnight incubation, the cells were washed with PBS and incubated with 50 μg/mL propidium iodide (Sigma Co.) and 50 μg/mL RNase A (Sigma Co.) in PBS at room temperature for 30 min. The fractions of cells in each phase of cell cycle were analyzed using FACScan flow cytometer and Cell Quest software (Becton Dickinson). Statistical Analysis. Data were expressed as means ( SD. Statistical analyses were performed using the Bonferroni t test method after ANOVA for multigroup comparison and the Student’s t test method for two group comparison, with p < 0.05 considered to be statistically significant. RESULTS AND DISCUSSION

Compound 2 (cohulupone) has been reported, but detailed spectroscopic data did not appear in the literature (15). The structure of 2 was confirmed by means of a spectroscopic method (Materials and Methods). The molecular formula of 3 was determined to be C30H44O6 by HREIMS (m/z 500.3137 [M]þ, Δ - 0.1 mmu). The IR spectrum of 3 implied the presence of hydroxy (3416 cm-1), carbonyl (1727 cm-1), conjugated carbonyl (1667 cm-1), and CdC (1620 cm-1) moieties. The 1H and 13C NMR spectra (Materials and Methods) of 3 resembled those of garsubelin D (16) except for a 2-(20 -hydroxypropyl)-3-hydroxy-2,3-dihydrofuran moiety. Thus, garcinielliptone P (3) was found to possess a bicyclic [3.3.1]nonane skeleton. The HMBC correlations between Me-20/C-21 and C-18

Article

J. Agric. Food Chem., Vol. 59, No. 1, 2011

411

Figure 9. Effect of 1 combined with cisplatin on the production of ROS in NTUB1 cells: 5 μM cisplatin þ 25 μM 1 (A), 5 μM cisplatin þ 50 μM 1 (B), 5 μM cisplatin þ 75 μM 1 (C), 10 μM cisplatin þ 25 μM 1 (D), 10 μM cisplatin þ 50 μM 1 (E), and 10 μM cisplatin þ 75 μM 1 (F), for 24 h. The amount of ROS was assayed by H2DCFDA staining. Each sampling measured the fluorescence intensity of region M1 of 3  105 cells cotreated via autofluorescence. The control cells were treated with medium. Results were repeated by independent experiments.

and Me-21/C-20 and C-18, and with C-19 being an oxygenated quaternary carbon, were used to establish that a 2-(20 -hydroxypropyl) unit is linked to C-18. Analysis of the 1H-1H COSY and 1H-13C HMQC spectra confirmed a bond between C-18 and C-17. In addition, the HMBC correlation of H-17/C-3 and the NOESY cross-peak between H-17/H-18 confirmed the presence of a 2β-(20 -hydroxypropyl)-2β-hydroxy-2,3-dihydrofuran moiety located at C-3 and C-4. Thus, garcinielliptone P (3) was characterized as 9,9-dimethyl-2R,8R-di(γ,γ-dimethylallyl-3,4-[2β-(20 hydroxypropyl)-2β-hydroxy-2,3-dihydrofuran]-6R-(1-oxo-2-methylpropyl)-8β-H-cis-bicyclo[3.3.1]nona-3-en-1,5-dione. The molecular formula of 4 was determined to be C25H22O10 by HRESIMS (m/z 505.1114 [M þ Na]þ, Δ þ 0.3 mmu), consistent with the 1H and 13C NMR data of this compound (Materials and Methods). The UV spectrum showed the presence of a xanthone moiety (λmax 212, 245, 265, and 312 nm) (17). The bathochromic shift induced by addition of AlCl3 suggested the presence of a hydroxy group at either C-1 or C-8. The IR absorption of 4 implied the presence of hydroxy (3443 cm-1) and conjugated carbonyl (1633 cm-1) groups. The 1H NMR spectrum displayed signals of a hydrogen-bonded hydroxy group [δ 13.03 (1H, s)]. The carbon signals of C-5 to C-8 and proton signals of H-5 to H-8 of 4 were found to be identical to those of the corresponding

signals of hyperielliptone HD (5) while carbon signals of hydroxymethyl to C-60 , C-100 to C-600 , OMe-12, and OMe-13, and proton signals of hydroxymethyl to H-60 , H-600 , H-200 , OMe-12, and OMe-13 were identical to those of the corresponding signals of hyperielliptone HC (5). Based on the above results, together with 4 (hyperielliptone HF) showing an additional methoxy carbon signal at δ 61.0, the structure of this compound was characterized as 50 -hydroxymethyl-60 -(400 -hydroxy-300 ,500 -dimethoxyphenyl)20 ,30 ;3,4-(1-hydroxy-2-methoxyxanthone)-10 ,40 -dioxane (4). The 2D NMR spectrum (Figure 2) of 4 also supported the structure assigned. In addition, the carbon signals C-3, C-4, hydroxymethyl, C-50 to C-60 , and C-100 were identical to those of corresponding carbon signals of 50 -demethoxycardensin G (18, 19). Thus, the orientation of the xanthone-phenylpropanoid ether linkage of 4 is assumed to be the same as that established for 50 -demethoxycardensin G. The optical inactivity and the trans relationship of the dioxane protons suggested that 4 was isolated as a racemic mixture of 50 R,60 R and 50 S,60 S enantiomers (20). A recent study demonstrated that phloroglucinols and triterpenoids have a significant antioxidant effect (6, 21). For evaluating the antioxidant effect of the isolated constituents, we used several methods, namely, inhibition of XO activity, 1,1-diphenyl2-picrylhydrazyl (DPPH) radical scavenging activity, and

412

J. Agric. Food Chem., Vol. 59, No. 1, 2011

Lin et al.

Figure 10. Flow cytometry analysis of cisplatin- and 1-treated NTUB1 cells. NTUB1 cells (3  105 cells per 6 cm dish) were treated in the absence of cisplatin or compound: control (A), 5 μM cisplatin (B), 10 μM cisplatin (C), 25 μM 1 (D), 50 μM 1 (E), 75 μM 1 (F), 5 μM cisplatin þ 25 μM 1 (G), 5 μM cisplatin þ 50 μM 1 (H), 5 μM cisplatin þ 75 μM 1 (I), 10 μM cisplatin þ 25 μM 1 (J), 10 μM cisplatin þ 50 μM 1 (K), and 10 μM cisplatin þ 75 μM 1 (L), for 24 h. At the indicated time, cells were stained with propidium iodide (PI), DNA contents were analyzed via flow cytometry, and the amount of apoptosis was measured by accumulation of sub-G1 DNA contents in the cells. The control cells were treated with medium. Results are representative of three independent experiments.

2,20 -azinobis(3-ethylbenzthiazoline-6-sulfonic acid) (ABTS) radicalcation scavenging activity. The antioxidant activities of 1-4, 5/5a (5), and allopurinol (positive control), a drug clinically prescribed for gout treatment, on XO activity were studied in vitro. As shown in Figure 3, compounds 1-3, 5/5a (5), and allopurinol inhibited XO activity in a concentration-dependent manner with an IC50 values of 130.2 ( 5.7, 165.2 ( 5.7, 48.1 ( 8.3, 42.1 ( 5.8 (5), and 3.2 ( 0.3 μM, respectively, while 4 did not display inhibitory effect on XO activity. The effect of antioxidant on DPPH is thought to be due to their hydrogen-donating activity (22). Compounds 4 and 5/5a exhibited weak DPPH•-scavenging activity

with 26.4 ( 1.3% and 4.5 ( 0.0% (5) inhibition at 300 μM while 1-3 did not display DPPH-scavenging activity. The positive control, tocopherol (TOC), displayed DPPH-scavenging activity with an IC50 value of 26.5 ( 0.3 μM. The ABTS radical-cation assay is an excellent method for determining the antioxidant activity of hydrogen-donating antioxidants and of chain-breaking antioxidants. As shown in Figure 4, compound 4 and positive control TOC exhibited ABTS radical-cation scavenging activity in a concentration-dependent manner with an IC50 value of 80.0 ( 4.0 and 47.6 ( 12.3 μM, respectively, while 2 revealed weak ABTS radical-cation scavenging activity with 25.7 ( 0.8% inhibition at

Article 150 μM. The antioxidants, such as pyrrolidine dithiocarbamate, epigallocatechin gallate, genistein, and vitamin E, showed synergistic cytotoxicity to PC-3 cells when combined with taxol (8). In addition, recently we have reported that a low dose of XO inhibitory terpenoid combined with cisplatin enhances the cytotoxic effect against NTUB1 cells (23) and 5/5a identified as a potent XO inhibitor (5). For continual evaluation of the cytotoxic effect and mechanism of induced cell death of XO inhibitors in vitro, the authors first selected 1 and 5/5a to examine the cytotoxicity against human NTUB1 cells. Cisplatin was used as positive control in the cytotoxic assay. As shown in Figures 5 and 6, 1 and 5/5a caused a significantly increased level of NTUB1 cell death in a concentration-dependent manner with an IC50 values of 44.1 ( 1.9 and 88.8 ( 10.3 μM, respectively. Cisplatin has been successfully used as a chemotherapeutic agent against malignant solid tumors in the head and neck region. However, there have been several side effects, such as nephrotoxicity and cisplatin resistance, when clinically used for treatment of cancers. Further, the authors wanted to evaluate if 1 or 5/5a in combination of cisplatin would enhance the cytotoxicity induced by cisplatin. As shown in Figure 7, treatment with 25, 50, and 75 μM 1 and 5 and 10 μM cisplatin alone caused 19%, 57%, and 95% and caused 63% and 72% cell killing relative to untreated control cells, respectively. The combination of 1 and cisplatin as the concentrations used in Figure 7 caused 58%, 70%, 87%, 83%, 99%, and 100% cell killing compared to untreated control cells, respectively, suggesting at least an additive and possibly a more than additive effect of 1 and cisplatin when compared with 1 or cisplatin alone. Treatment of different concentrations of 5/5a combined with 5 μM cisplatin enhanced the cell death but did not enhance the cell death induced by cisplatin (Figure 6). ROS induce programmed cell death, induce or suppress the expression of many genes, and activate cell signaling cascades (24). Expose of cells to 5 and 10 μM cisplatin and selective 1 (25, 50, and 75 μM) for 24 h, respectively, caused a significant increase in the intracellular amount of ROS as determined with the fluorescent dye, H2DCFDA, which preferentially detected intracellular ROS (Figure 8A-8F). In turn, exposure of cells to 25, 50, and 75 μM 1 combined with 5 or 10 μM cisplatin, respectively, for 24 h caused a significant increase in the intracellular level of ROS determined with that of the same fluorescent dye detected for 1 (Figure 9A-9F). ROS cause a wide range of cellular responses including from transient growth arrest to permanent growth arrest, apoptosis, or necrosis, depending on the amount of ROS. These responses allow to remove damage induced by ROS or allow cells to remove damaged cell (25). The effect of positive control cisplatin and different concentrations of 1 on cell cycle progression was determined by using fluorescence-activated cell sorting (FACS) analysis in propidium iodide-stained NTUB1 cells. As shown in Figure 10A-10F, treatment with 50 and 75 μM 1 for 24 h induced G1 phase arrest in a dose-dependent manner, accompanied by an increase in the level of apoptotic cell death. Treatment of cells with 50 and 75 μM 1 for 24 h yielded G1 phase arrest and significantly increased the level of ROS in cells. It clearly exhibited that the cell arrest and apoptosis induced by 1 were correlated with ROS. For further evaluation of mechanisms of combination of 1 and cisplatin, the effect of combination of 1 and cisplatin on cell cycle progression was also determined by using FACS analysis in propidium iodide-stained NTUB1 cells. As shown in Figure 10G-10L, treatment of 5 or 10 μM cisplatin combined with 25 μM 1, respectively, for 24 h induced G2/M and S phase arrest, accompanied by accumulation of cells in sub-G1 phase. Treatment of 5 or 10 μM cisplatin combined with 50 μM 1, respectively, for 24 h

J. Agric. Food Chem., Vol. 59, No. 1, 2011

413

induced G2/M arrest, accompanied by an increase in the level of apoptotic cell death. Treatment of 5 or 10 μM cisplatin combined with 75 μM 1, respectively, for 24 h induced G1 phase arrest, also accompanied by an increase in the level of apoptotic cell death. All inhibition of tubulin polymerization has been implicated in G2/M phase arrest in various cancer cell lines (26). The above result clearly indicated that a partial mechanism by which 25, 50, and 75 μM 1 combined with 5 or 10 μM cisplatin, respectively, mediated through generation of ROS in NTUB1 cells induced G2/M and S, G2/M, and G1 cell cycle arrest and apoptosis. It also suggested that treatment of different concentrations of 1 combined with 5 or 10 μM cisplatin, respectively, for 24 h induced different level of ROS and different cell phase arrest. In conclusion, compounds 1, a triterpenoid, and 5/5a, a phloroglucinol, exhibited weak cytotoxic activities against human NTUB1 cells. The combination of 50 and 75 μM 1 and 5 or 10 μM cisplatin enhanced the cell death induced by cisplatin while combination of different concentrations of 5/5a and 5 μM cisplatin did not enhance the cell death induced by cisplatin. Compound 1 or 1 combined with cisplatin mediated through generation of ROS in NTUB1 cells, induction of cell cycle arrest, and apoptosis. Compound 1 at 50 μM exhibited weaker cytotoxicity than that of 5 or 10 μM cisplatin and significantly enhanced cell death induced by cisplatin. It suggested that the combination of 50 μM 1 and 5 or 10 μM cisplatin or a low dose of cisplatin may enhance the therapeutic efficacy of cisplatin and reduce the side effect and drug resistance of cisplatin. LITERATURE CITED (1) Weng, J.-R.; Lin, C.-N.; Tsao, L.-T.; Wang, J.-P. Novel and antiinflammatory constituents of Garcinia subelliptica. Chem.;Eur. J. 2003, 9, 1958-1963. (2) Weng, J.-R.; Lin, C.-N.; Tsao, L.-T.; Wang, J.-P. Terpenoids with a new skeleton and novel triterpenoids with anti-inflammatory effects from Garcinia subelliptica. Chem.;Eur. J. 2003, 9, 5520-5527. (3) Lu, Y.-H.; Wei, B.-L.; Ko, H.-H.; Lin, C.-N. DNA strand-scission by phloroglucinols and lignans from heartwood of Garcinia subelliptica Merr. and Justicia plants. Phytochemistry 2008, 69, 225-233. (4) Wu, C.-C.; Lu, Y.-H.; Wei, B.-L.; Yang, S.-C.; Won, S.-J.; Lin, C.-N. Phloroglucinols with prooxidant activity from Garcinia subelliptica. J. Nat. Prod. 2008, 71, 246-250. (5) Wu, C.-C.; Yen, M.-H.; Yang, S.-C.; Lin, C.-N. Phloroglucinols with antioxidant activity and xanthonolignoids from the heartwood of Hypericum geminiflorum. J. Nat. Prod. 2008, 71, 1027-1031. (6) Lin., K.-W.; Huang., A.-M.; Tu, H.-Y.; Weng, J.-R.; Hour, T.-C.; Wei, B.-L.; Yang, S.-C.; Wang, J.-P.; Pu, Y.-S.; Lin, C.-N. Phloroglucinols inhibit chemical mediators and xanthine oxidase, and protect cisplatin-induced cell death by reducing reactive oxygen species in normal human urothelial and bladder cancer cells. J. Agric. Food Chem. 2009, 57, 8782-8787. (7) Maitraie, D.; Huang, C.-F.; Tu, H.-Y.; Liou, Y.-T.; Wei, B.-L.; Yang, S.-C.; Wang, J.-P.; Lin, C.-N. Synthesis, anti-inflammatory, and antioxidant activities of 18β-glycyrrhetinic acid derivatives as chemical mediators and xanthine oxidase inhibitors. Bioorg. Med. Chem. 2009, 17, 2785-2792. (8) Ping, S.-Y.; Hour, T.-C.; Lin, S.-R.; Yu, D.-S. Taxol synergizes with antioxidants in inhibiting hormal refractory prostate cancer cell growth. Urol. Oncol: Semin. Orig. Invest. 2010, 28, 170-179. (9) Simons, A. L.; Ahmad, I. M.; Mattson, D. M.; Dornfeld, K. J. 2-DexoyD-glucose combined with cisplatin enhances cytotoxicity via metabolic oxidative stress in human heed and neck cancer cells. Cancer Res. 2007, 67, 3364-3370. (10) Seo, S.; Tomita, Y.; Tori, K. Carbon-13 NMR spectra of urs-12-enes and application to structural assignments of components of Isodon japonicus HARA tissue cultures. Tetrahedron Lett. 1975, 7-10. (11) Liu, C.-H.; Yen, M.-H.; Tsang, S.-F.; Gan, K.-H.; Hsu, H.-Y.; Lin, C.-N. Antioxidant triterpenoids from the stems of Momordica charantia. Food Chem. 2010, 118, 751-756.

414

J. Agric. Food Chem., Vol. 59, No. 1, 2011

(12) Hour, T.-C.; Chen, J.; Huang, C.-Y.; Guan, J.-Y.; Lu, S.-H.; Hsieh, C.-Y.; Pu, Y.-S. Characterization of chemoresistance mechanisms in a series of cisplatin-resistant transitional carcinoma cell lines. Anticancer Res. 2000, 20, 3221-3225. (13) Pu, Y.-S.; Hour, T.-C.; Chen, J.; Huang, C.-Y.; Guan, J.-Y.; Lu, S.-H. Arsenic trioxide as a novel anticancer agent against human transitional carcinoma-characterizing its apoptotic pathway. Anticancer Drugs 2002, 13, 293-300. (14) Huang, A.-M.; Montagna, C.; Sharan, S.; Ni, Y.; Ried, T.; Sterneck, E. Loss of CCAAT/enhancer binding protein delta promotes chromosomal instability. Oncogene 2004, 23, 1549-1557. (15) Schmalreck, A. F.; Teuber, M.; Reininger, W.; Hartl, A. Structural features determining the antibiotic potencies of natural and synthetic hop bitter resins, their precursors and derivatives. Can. J. Microbiol. 1975, 21, 205-212. (16) Fukuyama, Y.; Minami, H.; Kuwayama, A. Garsubellins, polyisoprenylated phloroglucinol derivatives from Garcinia subelliptica. Phytochemistry 1998, 49, 853-857. (17) Scott, A. I. Interpretation of the Ultraviolet Spectra of Natural Products; Pergamon Press: London, 1964; p 158. (18) Sia, G.-L.; Bennett, G. J.; Harrison, L. J.; Sim, K.-Y. Minor xanthones from the bark of Crotoxylum cochinchinense. Phyotochemistry 1995, 38, 1521-1528. (19) Biemann, K. Tables of Spectral Data for Structure Determination of Organic Compounds; Springer-Verlag: New York, 1989; p C120. (20) Cardona, M. L.; Fernandez, M. I.; Pedro, J. R.; Seoane, E.; Vidal, R. Additional new xanthones and xanthonolignoids from Hypericum canariensis. J. Nat. Prod. 1986, 49, 95-100.

Lin et al. (21) Yin, M. C.; Chen, K. C. Nonenzymatic antioxidative and antiglycative effects of oleanolic acid and ursolic acid. J. Agric. Food Chem. 2007, 55, 7177-7181. (22) Mathew, S.; Abraham, T. E. In vitro antioxidant activity and scavenging effect of Cinnamom verum leaf extract assayed by different methodologied. Food Chem. Toxicol. 2006, 44, 198206. (23) Lin, C.-N. ; Huang, A.-M. ; Lin, K.- W. ; Hour, T.-C. ; Ko, H.-H. ; Yang, S.-C. ; Pu, Y.-S. Xanthine oxidase inhibitory terpenoids of Amentotaxus formosana protect cisplatin-induced cell death by reducing reactive oxygen species (ROS) in normal human urothelial and bladder cancer cells. Phytochemistry 2010, 71, 2140-2146. (24) Hancock, J. T.; Desikan, R.; Neill, S. J. Role of reactive oxygen species in cell signaling pathway. Biochem. Soc. Trans. 2001, 29 (Part 2), 345-350. (25) Boonstra, J.; Post, J. A. Molecular events associated with reactive oxygen species and cell cycle progression in mammalian cells. Gene 2004, 337, 1-13. (26) Tu, H.-Y.; Huang, A.-M.; Wei, B.-L.; Gan, K.-H.; Hour, T.-C.; Yang, S.-C.; Pu, Y.-S.; Lin, C.-N. Ursolic acid derivatives induce cell cycle arrest and apoptosis in NTUB1 cells associated with reactive oxygen species. Bioorg. Med. Chem. 2009, 17, 7265-7274. Received for review September 4, 2010. Revised manuscript received November 21, 2010. Accepted November 23, 2010. This work was supported by a grant from the National Science Council of the Republic of China (NSC 96-2628-B-037-012).