ARTICLE IN PRESS

Phytomedicine 14 (2007) 755762 www.elsevier.de/phymed

Antioxidant and hepatoprotective actions of medicinal herb, Terminalia catappa L. from Okinawa Island and its tannin corilagin
S. Kinoshitaa, Y. Inouea, S. Nakamaa, T. Ichibab, Y. Aniyaa,
a Laboratory of Functional and Molecular Pharmacology, Graduate School of Medicine, University of the Ryukyus, 207 Uehara, Nishihara, Okinawa 903-0215, Japan b Okinawa Industrial Technology Center, Okinawa 904-2234, Japan

Abstract
The antioxidant and hepatoprotective actions of Terminalia catappa L. collected from Okinawa Island were evaluated in vitro and in vivo using leaves extract and isolated antioxidants. A water extract of the leaves of T. catappa showed a strong radical scavenging action for 1,1-diphenyl-2-picrylhydrazyl and superoxide (O2 ) anion. Chebulagic acid and corilagin were isolated as the active components from T. catappa. Both antioxidants showed a strong scavenging action for O2 and peroxyl radicals and also inhibited reactive oxygen species production from leukocytes stimulated by phorbol-12-myristate acetate. Galactosamine (GalN, 600 mg/kg, s.c.,) and lipopolysaccharide (LPS, 0.5 mg/kg, i.p.)-induced hepatotoxicity of rats as seen by an elevation of serum alanine aminotransferase, aspartate aminotransferase and glutathione S-transferase (GST) activities was signicantly reduced when the herb extract or corilagin was given intraperitoneally to rats prior to GalN/LPS treatment. Increase of free radical formation and lipid peroxidation in mitochondria caused by GalN/LPS treatment were also decreased by pretreatment with the herb/ corilagin. In addition, apoptotic events such as DNA fragmentation and the increase in caspase-3 activity in the liver observed with GalN/LPS treatment were prevented by the pretreatment with the herb/corilagin. These results show that the extract of T. catappa and its antioxidant, corilagin are protective against GalN/LPS-induced liver injury through suppression of oxidative stress and apoptosis. r 2006 Elsevier GmbH. All rights reserved.
Keywords: Antioxidant; Corilagin; D-galactosamine; Terminalia catappa; Apoptosis

Introduction
Terminalia catappa L. is widely distributed in tropical and subtropical countries and the leaves of this plant has been used as a folk medicine for treating dermatitis and hepatitis in Asian countries (Lin and Kan, 1990; Perry, 1980). It has been reported that the extract of T. catappa leaves shows antioxidative, anti-inammatory and
Corresponding author. Tel.: +81 98 895 1251; fax: +81 98 895 1443. E-mail address: yaniya@med.u-ryukyu.ac.jp (Y. Aniya).

hepatoprotective actions (Gao et al., 2004; Lin et al., 1999; Tang et al., 2004) and contains hydrolysable tannins or triterpenoids (Gao et al., 2004; Lin et al., 2001; Tanaka et al., 1986). In Okinawa islands located between Taiwan and the mainland of Japan this plant also grows wild and seeds or young leaves of the plant have been rarely taken as food whereas it has not been known well as a folk medicine. Recently a chemopreventive action of the T. catappa on colon cancer in animal model was reported (Morioka et al, 2005). We also observed that the extract of T. catappa leaves shows the strongest action as a free

0944-7113/$ - see front matter r 2006 Elsevier GmbH. All rights reserved. doi:10.1016/j.phymed.2006.12.012

ARTICLE IN PRESS
756 S. Kinoshita et al. / Phytomedicine 14 (2007) 755762

radical scavenger among Okinawan medicinal herbs. Since constituents present in plants vary with environments of the plant, it was assumed that T. catappa growing in Okinawa islands may involve unique components. In the present study, we evaluated antioxidant and hepatoprotective actions of leaves extract of Okinawan T. catappa using D-galactosamine (GalN)/lipopolysaccharide (LPS)-induced liver injury of rats. Also we isolated antioxidant components from the extract of T. catappa and examined their antioxidant /hepatoprotective properties.

Aniya, 2000). Effects of the antioxidants on enzymatic/nonenzymatic lipid peroxidation was also measured according to the previous method (Gyam and Aniya, 2002).

Treatment of animals
Male SpragueDawley rats (300350 g, Seaku-Yoshitomi, Fukuoka, Japan) were divided into four groups randomly. First group received GalN (600 mg/kg) subcutaneously and LPS (0.5 mg/kg) intraperitoneally. Second group was given the T. catappa extract (1 mg/kg) intraperitoneally 1 and 15 h before GalN/LPS treatment. Third (control) and fourth groups were received water and the extract alone, respectively. In corilagin treatment, corilagin (1 mg/kg) was given intraperitoneally 1 and 15 h before GalN/LPS-treatment.

Materials and methods

Chemicals
Reduced glutathione (GSH) and lipopolysaccharide (LPS) were purchased from Sigma Chemicals (St. Louis, MO, USA). 1-Chloro-2,4-dinitrobenzene (CDNB), D-galactosamine (GalN), 1,1-diphenyl-2-picrylhydrazyl (DPPH) and 1,4-diazabicyclo [2,2,2]octane (DABCO) were from Wako Pure Chemicals (Osaka, Japan). Xanthine oxidase and 5,50 -dimethyl-1-pyrroline-N-oxide (DMPO) were from Boehringer Mannheim Gmbh (Mannheim, Germany) and Dojindo Laboratories (Kumamoto, Japan), respectively. Acetyl-Asp-Glu-ValAsp-4-methylcoumaryl-7-amide (AC-DEVD-MCA) was from Peptide Institute (Osaka, Japan). All reagents used were of analytical grade.

Preparation of the liver mitochondria and cytosol

Rats were sacriced by decapitation 24 h after GalN/ LPS treatment after an overnight starvation. The blood was collected from the stump and the serum was separated by centrifugation. The liver removed after perfusion in situ with ice-cold 1.15% potassium chloride solution containing a protease inhibitor, benzamidine (5 mM) was homogenized with 4 volumes of 0.03 M TrisHCl buffer (pH 7.0) including 0.25 M mannitol and 0.1 M potassium chloride (isolation buffer). The homogenate was centrifuged at 600g for 10 min at 4oC and the supernatant was further centrifuged at 15,000g for 30 min. The pellet thus obtained was washed twice by centrifugation and stored as mitochondria after suspension with the same buffer at 80 1C until use. On the other hand, the supernatant obtained by centrifugation at 15,000g was further centrifuged at 105,000g for 60 min at 4 1C and the resultant supernatant was used as cytosol. Animal care was in compliance with the Guidelines for Animal Experimentation of University of the Ryukyus.

Preparation of the herbal extract and isolation of antioxidant components

Dried cut leaves of T. catappa supplied by a company (Nakazen Co., Ltd.) in Okinawa were extracted with hot water (90 1C) for 1 h (1 g/10 ml). The extract was dried by a spray-dry method and was used as a crude extract for in vitro and in vivo experiments. Antioxidantive components from T. catappa were isolated and identied by similar method as reported previously (Aniya et al, 2002). Briey, 50% ethanol extract of dried leaves was chromatographed on a Toyopearl HW-40F column (Tosoh Cooperation, Tokyo, Japan). Fractions eluted from the column were subjected to a centrifugal partition chromatography (Miki-Engineering, Japan) and then fractionated by HPLC (Symmetry C18, Waters, USA). The structure of two compounds with a strong antioxidant action were identied by UV, 1H NMR and 13C NMR spectra. Antioxidant actions of the herb extract and its antioxidant components were evaluated by scavenging DPPH, superoxide anion (O2 ), peroxyl radicals and reactive oxygen species (ROS) from PMA-stimulated leukocytes as described previously (Myagmar and

Assay of serum and liver parameters

Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) activities in serum were measured using an assay kit (Kainos, Tokyo, Japan). Glutathione S-transferase (GST) activities in serum and liver cytosol were determined by using 1 mM CDNB and 1 mM GSH as substrates by the method of Habig et al. (1974). Lipid peroxide in the liver homogenate was evaluated by measuring thiobarbituric acid reactive substances (TBARS) (Gyam et al., 1999). Nitric oxide in serum and liver homogenate was measured as NO 2 and NO3 by using NO analyzer (Acom, Ltd., Tokyo, Japan).

ARTICLE IN PRESS
S. Kinoshita et al. / Phytomedicine 14 (2007) 755762 757

Caspase-3 activity in the supernatant after centrifugation at 15,000g as mentioned above was measured. The supernatant (100 ml) was incubated with AC-DEVDMCA (50 mM) in 0.1 M TrisHCl buffer (pH 8.0) in a total volume of 800 ml at room temperature for 3 min and then uorescence was measured at 370 nm exciting and 460 nm emission wavelengths. DNA fragmentation of the liver was examined by using the Kiagen genomic DNA purication kit (Kiagen Co. Ltd., Tokyo, Japan) followed by electrophoresis on a 2% agarose gel. Protein concentration in subcellular fractions was measured by the method of Lowry et al. (1951).

HO HO

OH

OH OH OH O O O O O OH R OH OH O

O O O HOOC HO O O H O

O OH

OH

R = H corilagin

Measurement of free radicals from mitochondria by electron spin resonance (ESR) spectrometer
Mitochondrial suspension (protein 500 mg) was incubated with 0.1% dodecyl maltoside, 5 mM glutamate, 5 mM malate and 100 mM succinate in 0.1 M potassium phosphate buffer (pH 7.4) in the presence of 2 mM NADH (5 ml) and 10 ml of DMPO in a total volume of 100 ml at 37 1C for 5 min and then the ESR signal was monitored. Free radical formation was evaluated by calculation of the relative peak height: dividing the rst peak height of a 4 line signals with that of MnO. Setting conditions of the ESR spectrometer (JES-FR30, JEOL) were as follows: power, 4 mW; magnetic eld center, 336.300 mT; sweep time, 1 min; modulation width, 0.1 mT; amplitude, 790; time constant, 0.1 s.

Chebulagic acid

Fig. 1. Structure of two antioxidants isolated from T. catappa collected from Okinawa Island.

and 16.4 mM (corilagin). Corilagin (IC50, 69 mM) inhibited ROS generation from PMA-stimulated leukocytes more than that of chebulagic acid (IC50, 154 mM). Both compounds also markedly inhibited enzymatic and non-enzymatic lipid peroxidation (IC50 values; 1.64.5 mM). We conrmed that chebulagic acid and corilagin are involved in both hot water extract and 50% ethanol extract (Fig. 2).

Hepatoprotective action of the herbal extract Statistical analysis

Data obtained from in vivo experiments are expressed as mean7SD. The statistical signicance was analyzed using t-test and p-values less than 0.05 were dened as signicant (*po0.05, **po0.01, ***po0.001 vs control; #po0.05, ##po0.01, ###po0.001 vs. Gal/ LPS treated). As shown in Fig. 3, serum ALT (18945% of control), AST (4439%) and GST (2889%) levels elevated by Gal/LPS treatment were signicantly decreased by pretreatment with T. catapa extract. The cytosolic GST, which is known to be released into serum in chemical-mediated liver injury (Aniya and Anders, 1985), activity was decreased to 76% in GalN/LPStreated rats and to 88% by the extract pretreatment. Lipid peroxide level in the liver homogenate (320%) increased by GalN/LPS was decreased by the extract (210%). Similarly, nitric oxide (NO) levels in the liver and serum were increased to 184% and to 170% of control by GalN/LPS treatment and decreased to 108% and 124% by the herb treatment, respectively. Serum AST and ALT levels of rats treated with the herb extract alone were similar to those of control rats (data not shown).

Results
Identication and properties of antioxidants from T. catappa
The hot water extract of T. catappa showed strong radical scavenging action and IC50 values (concentration with 50% scavenging action) for DPPH and O2 were 0.85 mg/ml and 0.2 mg/ml, respectively. We isolated two antioxidants from 50% ethanol extract and identied as chebulagic acid and corilagin (Fig. 1) yielding in 0.86% and 0.6%, respectively. These two antioxidants showed strong radical scavenging activity: IC50 values for O2 were 0.84 mM (chebulagic acid) and 0.32 mM (corilagin) and for peroxyl radical were 14.5 mM (chebulagic acid)

Effect of T. catappa extract on reactive oxygen species (ROS) generation and apoptosis induced by GalN/LPS treatment
The ESR signal with a 4-peak line was detected in mitochondria from GalN/LPS-treated rats whereas it

ARTICLE IN PRESS
758 S. Kinoshita et al. / Phytomedicine 14 (2007) 755762

Chebulagic acid
11.84

(1) 50% Ethanol extract

% 0 0.00

12.27

2.00

4.00

6.00

8.00

10.00

12.00
11.85 12.24

14.00

16.00

18.00

20.00

22.00

(2) Hot water extract

% 0 0.00 2.00 4.00 6.00 8.00 10.00

(3) Standard chebulagic acid

% 0 0.00

12.00 11.86

14.00

16.00

18.00

20.00

22.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

16.00

18.00

20.00

22.00

Corilagin (1) 50% Ethanol extract

% 0 0.00
3.44 6.32 11.03 10.42 9.65 9.49

17.33

2.00

4.00

6.00

8.00

10.00

12.00

14.00

16.00

18.00

20.00

22.00

(2) Hot water extract

% 0 0.00 2.00
3.63 6.42 7.86

11.03 9.51 9.68 17.21 17.36 21.27

8.83

4.00

6.00

8.00

10.00

12.00

14.00

16.00

18.00

20.00

22.00

(3) Standard corilagin

%

11.03

10.07

Time

0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

16.00

18.00

20.00

22.00

Fig. 2. HPLC chromatogram of chebulagic acid (a) and corilagin (b). Extract of T. catappa with hot water or 50% ethanol was applied on Nova-Pak C18 Radial-Pak HPLC column (Waters, Milford, MA). A ow rate of 3 ml/min was maintained while all analyses were conducted at ambient temperature. The mobile phase consisted of acetonitrile/water/acetic acid 90:9:1 (solvent A) and water/acetic acid 99:1 (solvent B). The gradient program was as follows: 0% A isocratic (1 min), 095% A linear gradient (18 min), 95% A (4 min). Analyses were carried out by single ion recording (SIR) on a Micromass ZQ quadrupole mass spectrometer, using m/z 953.1 for chebulagic acid and m/z 633.1 for corilagin (dwell times were 500 ms for each transitions, and cone voltages were 60 and 52 V, respectively). (a) Chebulagic acid; (1) 50% ethanol extract, (2) hot water extract, (3) standard chebulagic acid. (b) Corilagin; (1) 50% ethanol extract, (2) hot water extract, (3) standard corilagin.

was scarcely observed in mitochondria from rats pretreated with the herbal extract (Fig. 4). The ESR peak shape (the peak ratio of almost 1:2:2:1, g 2.005, aN abH 1.468 mT) indicates DMPO-OH adduct (hydroxyl radical). The radical peaks were diminished by incubation with DABCO, a singlet oxygen quencher, suggesting that singlet oxygen may be involved in the radical formation in mitochondria. As shown in Fig. 5, DNA fragmentation and marked increase in caspase 3 activity were detected in the livers of GalN/LPS-treated rats whereas they were scarcely observed after pretreatment with the herb extract.

Effect of corilagin on GalN/LPS-induced liver injury

As shown in Fig. 6, the increase in AST (7049%), ALT (24867%) and GST (2040%) activities in serum of GalN/LPS-treated rats was signicantly decreased by pretreatment of rats with corilagin. The cytosolic GST activity, decreased to 50% in GalN/LPS-treated liver, was recovered to control level by pretreatment with corilagin. Lipid peroxide level in liver homogenate and mitochondria, which was increased to 300% and 140% by GalN/LPS treatment, was reduced to 220% and 60% by corilagin pretreatment, respectively. Chromosomal DNA fragmentation and the increase in caspase 3

ARTICLE IN PRESS
S. Kinoshita et al. / Phytomedicine 14 (2007) 755762 759

25000 *** Activity (% of control) 20000 15000 10000 * 5000 # 0 AST ALT * ###

C GL TC+GL (% of control)

450 400 350 300 250 200 150 100 *** ### GST 50 0 cyt-GST liver-LPO liver-NO serum-NO *

Fig. 3. Effect of T. catappa pretreatment on serum and liver parameters of GalN/LPS-treated rats. Results show mean7SD for 35 rats. Activities in control are as follows: (a) AST; 187.3750.2 KU, ALT; 30.274.6 KU, serum GST; 0.06970.003 mmol/ml. (b) Cytosolic (cyt) GST; 0.9470.281 mmol/mg/min, liver LPO; TBARS 5.1073.28 nmol/mg, liver NO; 22.777.8 pmol/mg, serum NO; 15.6873.53 nmol/ml. C; control, GL; GalN/LPS, TC; T. catappa.

activity caused by GalN/LPS treatment were also reduced by corilagin pretreatment (Fig. 7). Thus it was demonstrated that corilagin is protective against GalN/ LPS-induced liver injury as well as T. catappa extract.

Discussion
In this study, antioxidant and hepatoprotective actions of leaves of T. catappa collected from Okinawa Island and its antioxidants, chebulagic acid and corilagin, were evaluated in vitro and in vivo. Water extract of T. catappa markedly scavenged DPPH and O2 radicals. Chebulagic acid/corilagin showed strong scavenging action of various radicals such as O2 , peroxyl radical or ROS from activated leukocytes and inhibited lipid peroxidation. Thus it was suggested that chebulagic acid and corilagin greatly contribute to strong antioxidant action of T. catappa from Okinawa. However, since various tannins including geraniin or triterpenoids such as ursolic acid and asiatic acid have been isolated from T. catappa collected from Taiwan (Gao et al., 2004; Lin et al., 2001; Tanaka et al., 1986), these components may contribute to the antioxidant action in the herb used in this study. We examined hepatoprotective action of leaves extract of T. catappa and corilagin against GalN/LPSinduced liver injury. Administration of subtoxic dose of GalN together with LPS has been used for preparing animal model of fulminant hepatic failure in which apoptosis of hepatocytes is the major culprit and is caspase dependent (Nakama et al., 2001; Gujral et al., 2003). In the present study, it was demonstrated that the extract of T. catappa and corilagin are protective against

GalN/LPS-induced liver toxicity as evidenced by the reversed serum AST, ALT and GST activities, by a decrease in lipid peroxide /nitric oxide levels and by prevention of apoptosis. It has been evidenced that excess generation of ROS in mitochondria leads to liver injury accompanied with apoptosis (Min ana et al., 2002; Carreras et al., 2004). In mitochondrial respiration chain, superoxide anion is generated at the site of complex I and III followed by dismutation to hydrogen peroxide which is consequently converted to hydroxyl radical in the presence of metal ions (Boveris, 1984). If the respiration chain is blocked, more ROS is generated (Wang et al., 2004). As shown in Fig. 4, ROS radical was detected in mitochondria of the livers of GalN/LPS-treated rats and was diminished by pretreatment with the herb extract. It is therefore suggested that the respiration chain in mitochondria is injured by GalN/LPS treatment followed by stimulation of ROS generation and the herb extract may recover the mitochondrial dysfunction by suppression of ROS generation. In consideration that ROS is also generated from macrophages like Kupffer cells in the liver by GalN/LPS-treatment, the herbal extract might scavenge the ROS resulting in prevention of liver injury. Furthermore, we observed that DNA fragmentation and an increase in caspase 3 activity caused by GalN/LPS treatment were suppressed by the pretreatment with T. catappa (Fig. 5). Since DNA fragmentation and an increase in caspase 3 activity are typical phenomena of apoptotic hepatocytes (Gujral et al., 2003) it is clear that the extract of T. catappa prevents the GalN/LPS induced apoptosis of hepatocytes. Similar hepatoprotective action against GalN/LPSderived liver injury was observed by pretreatment with

ARTICLE IN PRESS
760 S. Kinoshita et al. / Phytomedicine 14 (2007) 755762

Control

TC + GalN/LPS

GalN/LPS

DABCO (f:50mM)

Peak height

MnO

MnO

MnO

MnO

250 Relative peak height (% of control) 200 150 100 50 0 C

GalN/LPS TC+GalN//LPS

Fig. 4. ROS generation from mitochondria: (a) ESR spectra and (b) relative peak height of the radical. TC: T. catappa; DABCO: 1,4-diazabicyclo [2,2,2] octane.

corilagin (Figs. 6 and 7). The fact that corilagin is involved in hot water and 50% ethanol extracts (Fig. 2) suggest that corilagin attributes, at least, to the hepatoprotective action of T. catappa. Although we examined the hepatoprotective effect of chebulagic acid, the same dose of chebulagic acid was not hepatoprotective (data not shown). More studies are needed to conrm the action of chebulagic acid. Hepatoprotective actions of T. catappa collected from Taiwan or of antioxidants isolated from the herb have also been reported (Gao et al., 2004; Tang et al., 2004). However, there is no report on corilagin. Thus we rstly

demonstrated that corilagin prevents GalN/LPS-induced liver injury through suppression of ROS generation and inhibition of apoptosis. In summary, T. catappa collected from Okinawa Island has a strong antioxidant action and is protective against GalN/LPS-induced liver injury through suppression of ROS generation followed by inhibition of apoptosis. Chebulagic acid and corilagin, identied as main antioxidants, showed strong scavenging action for various radicals. Corilagin contributes, at least, to the hepatoprotective action of T. catappa.

ARTICLE IN PRESS
S. Kinoshita et al. / Phytomedicine 14 (2007) 755762 761

4.5 4.0 Caspase-3 (pmol/mg/min) 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0 C GL TC+GL M C GL TC+GL

Fig. 5. Effect of T. catappa on apoptosis of livers of GalN/LPS-treated rats. (a) Caspase 3 activity; (b) DNA fragmentation. M: 500 bp ladder marker, GL: GalN/LPS, TC: T. catappa.

a
Activity (% of control)

30000 25000 20000 15000

**

C GL CR+GL

b
(% of control)

400 350 300 250 200 150 100 50 0 ** ##

**

## 10000 5000 0 AST

ALT

*** ## ** # GST

cytGST

liver-LPO

Fig. 6. Effect of corilagin (CR) on serum and liver parameters of GalN/LPS (GL)-treated rats. Results show mean7SD for 35 rats. The activities in control are as follows. (a) AST: 143.8719.5 KU, ALT: 27.370.9 KU, serum GST: 0.05170.016 mmol/ml and (b) cytosolic (cyt) GST: 1.12170.239 mmol/mg/min, liver-LPO: TBARS: 4.270.53 nmol/mg.

a
8.0 Caspase-3 (pmol/mg/min) 7.0 6.0 5.0 4.0 3.0 2.0 1.0 0.0 C GL CR+GL # *

GL

CR+GL

Fig. 7. Effect of corilagin on apoptosis of livers of GalN/LPS-treated rats. (a) Caspase 3 activity and (b) DNA fragmentation. M: 500 bp ladder marker, GL: GalN/LPS, CR: corilagin.

Acknowledgments
We wish to thank Mr. K. Nakamoto of Nakazen Company for kind supply of the herb, Terminalia catappa

and Ms. N. Murayama for typing the manuscript. This study was partially supported by a Grant-in Aid from Okinawa Promotion Program for Cooperative Research between Industry, the Academic World and Administration.

ARTICLE IN PRESS
762 S. Kinoshita et al. / Phytomedicine 14 (2007) 755762

References
Aniya, Y., Anders, M.W., 1985. Alteration of hepatic glutathione S-transferase and release into serum after treatment with brombenzene, carbon tetrachloride, or dimethylnitrosoamine. Biochem. Pharmacol. 34, 42394244. Aniya, Y., Miyagi, C., Nakandakari, A., Kamiya, S., Imaizumi, N., Ichiba, T., 2002. Free radical scavenging action of the medicinal herb Limonium wrightii from the Okinawa islands. Phytomedicine 9, 239244. Boveris, A., 1984. Determination of the production of superoxide radicals and hydrogen peroxide in mitochondria. Methods Enzymol. 105, 429435. Carreras, M.C., Franco, M.C., Peralta, J.G., Poderoso, J.J., 2004. Nitric oxide, complex I, and the modulation of mitochondrial reactive species in biology and disease. Mol. Aspects Med. 25, 125139. Gao, J., Tang, X., Dou, H., Fan, Y., Zhao, X., Xu, Q., 2004. Hepatoprotective activity of Terminalia catappa L. leaves and its two triterpenoids. J. Pharm. Pharmacol. 56, 14491455. Gujral, J.S., Farhood, A., Jaeschke, H., 2003. Oncotic necrosis and caspase-dependent apoptosis during galactosamineinduced liver injury in rats. Toxicol. Appl. Pharmacol. 190, 3746. Gyam, M.A., Aniya, Y., 2002. Antioxidant properties of Thonningianin A, isolated from the Arican medicinal herb, Thonningia sangunea. Biochem. Pharmacol. 63, 17251737. Gyam, M.A., Yonamine, M., Aniya, Y., 1999. Free radical scavenging action of medicinal herbs from Ghana Thonningia sanguinea on experimentally-induced liver injuries. Gen. Pharmacol. 32, 661667. Habig, W.H., Pabst, M.J., Jakoby, W.B., 1974. Glutathione S-transferase. J. Biol. Chem. 249, 71307139. Lin, C.C., Kan, W.S., 1990. Medical plants used for the treatment of hepatitis in Taiwan. Am. J. Chin. Med. 18, 3843. Lin, C.C., Hsu, Y.F., Lin, T.C., 1999. Effects of punicalagin and punicalin on carrageenan-induced inammation in rats. Am. J. Chin. Med. 27, 371376.

Lin, C.C., Hsu, Y.F., Lin, T.C., 2001. Antioxidant and free radical scavenging effects of the tannins of Terminalia catappa L. Anticancer Res 21, 237243. Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265275. mez-Cambronero, L., Lloret, A., Pallardo , Min ana, J.B., Go F.V., Del Olmo, J., Escudero, A., Rodrigo, J.M., Pell n, A., Vin a, J.R., Vin a, J., Sastre, J., 2002. Mitochondrial oxidative stress and CD95 ligand: a dual mechanism for hepatocyte apoptosis in chronic alcoholism. Hepatology 35, 12051214. Morioka, T., Suzui, M., Nabandith, V., Inamine, M., Aniya, Y., Nakayama, T., Ichiba, T., Yoshimi, N., 2005. Moding effects of Terminalia catappa on azoxiymethane-induced colon carcinogenesis in male F344 rats. Eur. J. Cancer Prev. 14, 101105. Myagmar, B.-E., Aniya, Y., 2000. Free radical scavenging action of medicinal herbs from Mongolia. Phytomedicine 7, 221229. Nakama, T., Hirono, S., Moriuchi, A., Hasuike, S., Nagata, K., Hori, T., Ido, A., Hayashi, K., Tsubouchi, H., 2001. Etoposide prevents apoptosis in mouse liver with Dgalactosamine/lipopolysaccharide-induced fulminant hepatic failure resulting in reduction of lethality. Hepatology 33, 14411450. Perry, L.M., 1980. In: Perry, L.M. (Ed.), Medicinal Plants of East and Southeast AsiaAttributed Properties and Uses. The Massachusetts Institute of Technology Press, p. 80. Tanaka, T., Nonaka, G.I., Nishioka, I., 1986. Tannins and related compounds. XLII. Isolation and characterization of four new hydrolysable tannins, teravins A and B, tergallagin and tercatain from the leaves of Terminalia catappa L. Chem. Pharm. Bull. 34, 10391049. Tang, X.H., Gao, L., Gao, J., Fan, Y.M., Xu, L.Z., Zhao, X.N., Xu, Q., 2004. Mechanisms of hepatoprotection of Terminalia catappa L. extract on D-galactosamine-induced liver damage. Am. J. Chin. Med. 2, 509519. Wang, Y., Fang, J., Leonard, S.S., Rao, K.M., 2004. Cadmium inhibits the electron transfer chain and induces reactive oxygen species. Free Radic. Biol. Med. 36, 14341443.