Protective Effects of Macelignan on Cisplatin-Induced Hepatotoxicity Is Associated with JNK Activation

February 2008 Biol. Pharm. Bull. 31(2) 273—277 (2008) 273 Protective Effects of Macelignan on Cisplatin-Induced Hepatotoxicity Is Associated with JNK Activation Jong Hee SOHN, Kyu Lee HAN, Jeong-Hwan KIM, Yaya RUKAYADI, and Jae-Kwan HWANG* Department of Biotechnology, Yonsei University; 134 Shinchon-dong, Seodaemun-gu, Seoul 120–749, Korea. Received June 13, 2007; accepted October 11, 2007 Cisplatin is one of the most effective antineoplastic drugs, but it has undesirable side effects such as hepato- toxicity at high doses. This study investigated the protective effect of macelignan, isolated from Myristica fragrans HOUTT. (nutmeg), against cisplatin-induced hepatotoxicity and the possible mechanisms involved in these effects in mice. Pretreatment with macelignan for 4 d significantly prevented the increased serum enzymatic activities of alanine and aspartate aminotransferase in a dose-dependent manner. The results also showed that the protective effects of macelignan on cisplatin-induced hepatotoxicity may be associated with the mitogen activated protein kinase (MAPK) signaling pathway. Cisplatin-induced phosphorylation of c-Jun N-terminal kinase1/2 (JNK1/2) and extracellular signal-regulated kinase1/2 (ERK1/2) was abrogated by pretreatment with macelignan, however, that of p38 was not significantly affected. It was also found that macelignan attenuated the expression of phos- phorylated c-Jun in cisplatin-treated mice. Accordingly, it is suggested that the hepatoprotective effects of macelignan could be related to activation of the MAPK signaling pathway, especially JNK and c-Jun, its sub- strate. The present findings suggest that co-treatment of cisplatin with macelignan may provide more advantage than cisplatin treatment alone in cancer therapy. Key words macelignan; cisplatin; hepatotoxicity; mitogen-activated protein kinase; c-Jun N-terminal kinase; c-Jun In the clinical field, the most important form of toxic he- The objective of the current study was to examine the pro- patic injury is caused by therapeutic agents. Drug-induced tective effects of macelignan on cisplatin-induced hepatotox- liver injury is the most frequent reason for the withdrawal of icity. For this purpose, we evaluated the levels of serum ala- an approved drug from the market, and it is also responsible nine aminotransferase (ALT) and aspartate aminotransferase for more than 50% of cases of acute liver failure in the (AST). Several studies have reported that cisplatin activity is United States.1) More than 600 drugs have been associated related to the cascades of mitogen-activated protein kinases with hepatotoxicity, and there are now a number of reports (MAPKs) including c-Jun N-terminal kinases (JNKs, also re- that reactive metabolites formed from drugs such as aceta- ferred to as stress-activated protein kinases, SAPK), extracel- minophen, tamoxifen, diclofenac, and troglitazone are known lular signal-regulated kinases (EKRs), and p38 kinases.18,19) to cause hepatotoxicity.2) Accordingly, we examined whether the protective activity of Cisplatin (cis-diamminedichloroplatinum[II]) is one of the macelignan against cisplatin-induced hepatotoxicity is asso- most effective antineoplastic drugs, and it is particularly used ciated with the mitogen-activated protein kinase (MAPK) for the treatment of ovarian, testicular and head and neck signaling pathway. cancers.3,4) In spite of its chemotherapeutic activity, various adverse effects, including renal dysfunction, nausea and MATERIALS AND METHODS vomiting, myelosuppression, ototoxicity, and nephrotoxicity, associated with its clinical use are well known.5) Among Isolation of Macelignan Macelignan (Fig. 1) was iso- them, nephrotoxicity is frequent and a major limitation to the lated from Myristica fragrans HOUTT. (Myristicaceae) as de- use of this drug. Several mechanisms, including oxidative scribed previously.20) Dried seed kernels (100 g) of Myristica stress, inflammation, genotoxic damage, and cell cycle arrest, fragrans were ground and extracted twice with 75% aqueous have been examined in cisplatin-induced nephrotoxicity.6—8) methanol (400 ml, v/v) for 24 h at room temperature. The However, cisplatin-induced liver toxicity and the mechanisms methanol extract was concentrated, frozen, lyophilized (7 g) have been little studied, except for some clinical reports.9—11) and further fractionated successively with ethyl acetate, n-bu- Myristica fragrans HOUTT. (Myristicaceae) is a perennial tanol and water. Each fraction was evaporated and dried herb native to Indonesia and cultivated in South Africa, the Molucca Islands, India and other tropical areas. Its fruits, commonly known as mace (outer husk) or nutmeg (inner seed or kernel), has been used traditionally for spice and also possesses carminative, astringent, hypolipidaemic, antithrombotic, antiplatelet aggregation, antifungal, aphro- disiac, anxiogenic, antidiarrheal and anti-inflammatory activ- ities.12—14) Macelignan (Fig. 1) isolated from M. fragrans has been reported to have antioxidant activity and cause alter- ation in hepatic enzyme activities,15—17) however, the protec- tive effect of macelignan against hepatotoxicity has not yet been examined. Fig. 1. Chemical Structure of Macelignan ∗ To whom correspondence should be addressed. e-mail: [email protected] © 2008 Pharmaceutical Society of Japan 274 Vol. 31, No. 2 under reduced pressure (ethyl acetate fraction 4.2 g, butanol EDTA, 1 mM phenylmethanesulphonyl fluoride (PMSF)) and fraction 0.7 g, water fraction 2.1 g). Further separation was incubated on ice for 20 min. The mixture was then cen- performed using silica gel column chromatography (70—230 trifuged, and the supernatant was used to determine the con- mesh, Merck & Co., Whitehouse Station, NJ, U.S.A.) by centration of protein. Fifty micrograms of total cellular pro- eluting with n-hexane : ethyl acetate solution (10 : 1, v/v), and tein was subjected to electrophoresis on SDS-polyacrylamide 10 ml volumes of eluant were collected in test tubes. The col- gels and transferred to nitrocellulose membranes. Nonspe- lected tubes were divided into six fractions (Fr. I—Fr. VI) cific binding was blocked overnight with 5% skim milk in following silica TLC (thin layer chromatography; 60 F254, Tris-buffered saline Tween-20 (TBST) (10 mM Tris, pH 7.5, Merck & Co., Whitehouse Station, NJ, U.S.A.). Fr. III, pro- 100 mM NaCl, 0.1% Tween 20) at 4 °C, and then blots were viding considerable inhibitory activity against S. mutans, was incubated with 1 : 1000 diluted primary antibody for 2 h. further separated through a silica gel column using n- After washing with TBST, the membranes were probed with hexane : ethyl acetate (20 : 1, v/v), yielding Fr. III-B (0.52 g). 1 : 2000 secondary antibody for 1 h and washed with TBST Fr. III-B was eluted with 80% methanol using Rp-18 column three times. Immunoblots were visualized using a chemilu- chromatography (LiChropep, 25—40 m m, Merck & Co., minescent reagent. The densities of the bands were measured Whitehouse Station, NJ, U.S.A.), and Fr. III-B-2 (0.5 g) was using an RFLPscan version 2.1 software program (Scanalyt- finally obtained as a single compound. ics Inc., Fairfax, VA, U.S.A.). Chemicals Cisplatin (cis-diamminedichloroplatinum Statistics All experiments used 6—8 animals per treat- [II]), curcumin, and corn oil were purchased from Sigma- ment group, and the results are presented as the mean
stan- Aldrich Co. (St. Louis, MO, U.S.A.). Antibodies used in this dard deviation (S.D.). Statistical analysis was performed experiment were obtained from Santa Cruz Biotechnology using one-way analysis of variance (ANOVA) with subse- Inc. (Santa Cruz, CA, U.S.A.), and ECL solution was pur- quent post hoc comparisons by Scheffé’s test (SPSS 12.0). chased from Amersham Pharmacia Biotech (Piscataway, NJ, The criterion for statistical significance is expressed as U.S.A.). * p0.05, ** p0.01. Animals and Treatment Male ICR mice weighing 25
5 g (5 weeks old) were purchased from Samtako Bio RESULTS Korea (Seoul, Korea). The animals were allowed free access to normal standard chow diet and tap water, and maintained Effect of Macelignan on Cisplatin-Induced Hepatotoxi- in controlled conditions of 25
2 °C, 55
5% relative humid- city The protective effect of macelignan on cisplatin-in- ity with a 12 h dark/light cycle. The protocol for mice main- duced hepatotoxicity was examined by quantifying the levels tenance was approved by the Committee for the Care and of serum ALT and AST. In cisplatin-treated mice, serum lev- Use of Laboratory Animals at Yonsei University, according els of ALT and AST were increased to 494% and 333% com- to the Guide for Animal Experiments edited by the Korean pared to the normal group (Fig. 2). However, pretreatment Academy Medical Association. Mice were kept in this envi- with macelignan was protective, as indicated by significant ronment for 1 week before an experiment. Macelignan in reduction of ALT and AST activities, and showed higher ac- corn oil (25 or 50 mg/kg per day) was administered orally for tivity than that with curcumin (50 mg/kg). In contrast, there four consecutive days. Three hours after the final treatment were no significant differences in the activities of ALT and with macelignan, a single dose of cisplatin (45 mg/kg, in AST in mice treated with macelignan (50 mg/kg) alone when PBS), which has been reported to induce hepatotoxicity in compared with control. This indicates that macelignan itself mice without lethality,21) was injected intraperitoneally. Corn does not have a hepatotoxic effect. oil and PBS were used as vehicle control, and the positive Effect of Macelignan on Hepatic Morphological group animals were given cisplatin with curcumin Changes of Cisplatin-Induced Hepatotoxicity Histologi- (50 mg/kg). Sixteen hours after administrating cisplatin, all cal effects were also examined in this study. In contrast to animals were killed with CO2 anesthesia, and blood and liver cisplatin-untreated control mice (Fig. 3a) and macelignan samples were collected. only treated mice (Fig. 3b), cisplatin-treated mice showed se- Determination of Serum Biochemical Parameters vere centrilobular necrosis and lymphocyte infiltration (Fig. Blood samples were kept at room temperature for 1 h, and 3c). Pretreatment with macelignan strongly prevented the de- serum was prepared by centrifugation at 4000 rpm for 15 min velopment of severe hepatic lesions induced by cisplatin, and stored at 70 °C until analysis. The levels of serum ALT with marked reductions in necrosis (Fig. 3d). These results and AST were measured using an automatic photometer were consistent with those of serum hepatotoxicity indices (PRIME, BPC BioSed S.r.l., Italy). shown in Fig. 2. Histopathological Examination Portions of the liver Effect of Macelignan on Cisplatin-Mediated Activation were fixed in 10% formalin and then embedded in paraffin. of MAP Kinases and Transcription Factors To investi- Five micrometer thick microtome sections were prepared gate whether the protective effect of macelignan is related to from each liver sample and stained with hematoxylin–eosin. MAP kinase signaling, protein samples obtained from mouse The sections were examined for the pathological findings of liver were subjected to Western blot analysis. As shown in hepatotoxicity such as centrilobular necrosis and lymphocyte Fig. 4, although the unphosphorylated form of JNK1/2 was infiltration. not changed by cisplatin treatment, its administration led to Western Blot Analysis Powdered liver tissues under liq- an increase in the level of phosphorylated JNK1/2. However, uid nitrogen were homogenized with lysis buffer (50 mM pretreatment with macelignan before cisplatin injection sig- Tris–HCl, pH 7.4, 150 mM NaCl, 0.1% sodium dodecyl sul- nificantly suppressed JNK1/2 activation, and this activity of fate (SDS), 1.0% Triton X-100, 0.25% deoxycholate, 1 mM macelignan was almost comparable to that of curcumin at the February 2008 275 macelignan. Moreover, phosphorylated JNK1/2, ERK1/2 and p38 were not affected when treated with macelignan alone (data not shown). Based on the above data, the protein expression levels of AP-1 subunit, c-Jun and c-Fos were examined. The results showed that expression of phosphorylated c-Jun was in- creased in cisplatin-treated mice, although unphosphorylated c-Jun and c-Fos production were not changed by administra- tion of cisplatin or macelignan (Fig. 4). However, increased phosphorylated c-Jun expression was highly attenuated by treatment with macelignan. In addition, macelignan pre- sented more efficacious activity than curcumin at the same dose. DISCUSSION Clinical evidence of cisplatin-induced hepatic injury has been demonstrated by elevated activities of serum amino- transferase including ALT and AST.9) The rise in levels of serum ALT and AST has been attributed to the damaged structural integrity of the liver, because these are normally located in the cytoplasm, and are released into the circulation after hepatic damage.22) Therefore, the protective effect of macelignan on cisplatin-induced hepatotoxicity was exam- ined by quantifying the levels of ALT and AST. As shown in Fig. 2, pretreatment with macelignan was protective, as indi- Fig. 2. Effect of Macelignan on Cisplatin-Induced Hepatotoxicity cated by the significant reductions in ALT and AST activi- Mice were pretreated with macelignan or curcumin orally (25 or 50 mg/kg, in corn ties. Histopathological analysis of liver sections indicated a oil) for 4 d. Three hours after the last administration, the mice were injected with cis- platin (45 mg/kg, i.p.). Sixteen hours after the injection, blood was collected and serum moderate centrilobular necrosis and lymphocytic infiltration levels of ALT and AST were measured. Each value is the mean
S.D. (n6—8). in mice treated with macelignan plus cisplatin compared ## p0.01 vs. control, ∗ p0.05, ∗∗ p0.01 vs. cisplatin-treated group. with those treated with cisplatin alone (Fig. 3). The cytotoxic action of cisplatin is mediated by its interac- tion with DNA to form DNA adducts, which activate several signal transduction pathways including MAPK and culmi- nate in the activation of apoptosis.23) The MAPK signaling pathways play important roles in regulating gene expression, cell proliferation, cell motility, and cell survival and death in eukaryotic cells. Three distinct groups of MAPKs have been identified in mammalian cells: c-Jun N-terminal kinase (JNKs), extracellular-regulated kinase (ERKs), and p38.24) A major role for MAPKs is to regulate the activities of tran- scription factor and transmit extracellular signals to the nu- cleus, which induce target gene expression. Numerous mam- malian transcription factors and their coregulatory proteins are known as targets of the MAPK cascades.25) A previous study has demonstrated that MAPK members can be activated following exposure of tumor cells to cis- platin.26) It has been well documented that curcumin has a hepatoprotective effect27—29) and is an inhibitor of JNK in he- patocytes.30,31) In this study, cisplatin induced expression of Fig. 3. Hepatoprotective Effect of Macelignan against Cisplatin-Induced phosphorylated JNK1/2 and ERK1/2 in mice liver, however, Acute Hepatotoxicity in Mice pretreatment with macelignan before cisplatin injection sup- Liver sections were stained with hematoxylin–eosin. (a) Normal; (b) macelignan pressed JNK and ERK activation more than curcumin (Fig. (50 mg/kg); (c) cisplatin; (d) macelignan (50 mg/kg)cisplatin. Magnification 200. 4). Cisplatin-induced kinase activation and its inhibition by same dose (50 mg/kg). Cisplatin also induced expression of macelignan were apparent in JNK1/2 signaling. JNK and phosphorylated ERK1/2. These effects were slightly attenu- ERK activation in cisplatin-treated cells has been reported, ated by macelignan, but the inhibitory activity was not appar- however, their roles are still controversial. Most studies have ent compared to that in JNKs. The protein levels of unphos- shown that activation of JNK is associated with induction of phorylated and phosphorylated p38 were not affected by ei- cell death.32) In contrast, other studies have suggested that ther cisplatin or a combination treatment of cisplatin with JNK signaling plays a role in enhanced cell survival of cis- 276 Vol. 31, No. 2 Fig. 4. Protein Expression Levels of MAP Kinases, c-Jun and c-Fos The protein levels of unphosphorylated and phosphorylated forms of MAPKs, c-Jun and c-Fos were measured in cytosolic proteins by Western blot analysis using specific anti- bodies. The relative intensity of the bands is expressed as the ratio of treated to untreated samples. platin-treated cells.33) JNK activation associated with cell fate response to genotoxic stress and pro-inflammatory cytokines, is dependent on the type of stimuli and circumstances, but it and phosphorylation of c-Jun enhances its transcriptional ac- seems clear that JNK is activated in cells treated with anti- tivity.37) c-Jun is necessary for the induction of apoptosis in neoplasic agents, such as cisplatin, vincristin or adriamycin, cells exposed to genotoxic stress or UV radiation.38,39) On the although its role on apoptosis is not well established.34) Like other hand, some previous studies have showed that c-Jun is JNK, ERK is also activated by cisplatin in a variety of cell rapidly activated in response to mitogens, indicating that c- lines. According to previous studies, ERK activation is asso- Jun probably has a function in the control of cell prolifera- ciated with either inducing apoptosis or antagonizing apopto- tion.40,41) Balance between the pro-apoptotic and anti-apop- sis of cisplatin-treated cells.35,36) Both effects of survival and totic target gene expression determines whether the final out- death signaling mediated through MAPK are possible, and come will be cell survival or cell death. This balance may be this phenomenon could reflect differences in cell context or dependent on the type and duration of stimulus, as well as on the extent of DNA damage.23) the activation of other transcription factors.42) It is well known that MAPK signaling pathways regulate It is reported that persistent activation of JNK can be in- expression of the AP-1 family members c-Jun and c-Fos.37) volved in cisplatin-induced apoptosis, and c-Jun, the main AP-1 transcription factors could be involved in both the in- substrate of JNK is necessary for induction of apoptosis in duction and prevention of apoptosis. c-Jun is a central com- response to cisplatin.43,44) In this study, the administration of ponent of all AP-1 complexes, which act often as transcrip- cisplatin led to the activation of JNKs, and its induction was tional activators. MAPKs such as JNK, ERK5, and p38 can significantly attenuated by macelignan treatment. Pretreat- regulate the expression levels of genes encoding these tran- ment with macelignan also inhibited c-Jun activation in cis- scription factors by controlling the c-jun promoter depending platin-treated mice. These results indicate that macelignan at- on the stimulus.24) c-Jun is targeted by the JNK pathway in tenuates activation of JNK and c-Jun caused by cisplatin, February 2008 277 which prevents hepatocyte apoptosis. 17) Woo W. S., Shin K. H., Wagner H., Lotter H., Phytochemistry, 26, Like c-Jun, the activity of c-Fos, another component of 1542—1543 (1987). 18) Kim Y. K., Kim H. J., Kwon C. H., Kim J. H., Woo J. S., Jung J. S., AP-1, can be regulated by JNK signaling pathways.45) Induc- Kim J. M., J. Appl. Toxicol., 25, 374—382 (2005). tion of c-Fos is important for AP-1 activity because Jun/Fos 19) Mansouri A., Ridgway L. D., Korapati A. L., Zhang Q., Tian L., Wang heterodimers are more stable than Jun/Jun homodimers.46) Y., Siddik Z. H., Mills G. B., Claret F. X., J. Biol. Chem., 278, However, in this study, the protein expression of c-Fos was 19245—19256 (2003). not changed in cisplatin-treated mice liver. Considering other 20) Chung J. Y., Choo J. H., Lee M. H., Hwang J. K., Phytomedicine, 13, 261—266 (2005). reports about AP-1 activity, several mechanisms could be in- 21) Liu J., Liu Y., Habeebu S. S., Klaassen C. D., Toxicol. Appl. Pharma- volved in regulating AP-1 activity. The investigation of a col., 149, 24—31 (1998). more exact mechanism about JNK signaling and its sub- 22) Vermeulen N. P. E., Bessems J. G. M., van de Straat R., Drug Metab. strates affected by treating cisplatin and macelignan is Rev., 24, 367—407 (1992). 23) Siddik Z. H., Oncogene, 22, 7265—7279 (2003). needed. 24) Yang S. H., Sharrocks A. D., Whitmarsh A. J., Gene, 27, 3—21 In conclusion, macelignan reduced cisplatin-induced hepa- (2003). totoxicity in mice, and its protective activity might be related 25) Seger R., Krebs E. G., FASEB J., 9, 726—735 (1995). to the JNK signaling pathway. Phosphorylated JNKs and c- 26) Wang X., Martindale J. L., Holbrook N. J., J. Biol. Chem., 275, Jun, its substrate, were strongly induced in cisplatin-treated 39435—39443 (2000). 27) Kiso Y., Suzuki Y., Watanabe N., Oshima Y., Hikino H., Planta Med., mice, and these inductions were abrogated by administration 49, 185—187 (1983). with macelignan. The present findings suggest that macelig- 28) Donatus I. A., Sardjoko L., Vermeulen N. P., Biochem. Pharmacol., nan could be used as a protective agent against various hepa- 39, 1869—1875 (1990). totoxins such as cisplatin, acetaminophen, tamoxifen and 29) Park E. J., Jeon C. H., Ko G., Kim J., Sohn D. H., J. Pharm. Pharma- azathioprine which are known to activate JNK.47—49) Cis- col., 52, 437—440 (2000). 30) Zimmerman T. L., Thevananther S., Ghose R., Burns A. R., Karpen S. platin has been shown to induce acute renal failure in mice J., J. Biol. Chem., 281, 15434—15440 (2006). and humans.50) Further studies are necessary to confirm 31) Li D., Zimmerman T. L., Thevananther S., Lee H. Y., Kurie J. M., whether macelignan works in kidney or other tissues. It is an- Karpen S. J., J. Biol. Chem., 277, 31416—31422 (2002). ticipated that the clinical application of macelignan in cancer 32) Sanchez-Perez I., Murguia J. R., Perona R., Oncogene, 16, 533—540 therapy could provide more advantages to reducing side ef- (1998). 33) Potapova O., Haghighi A., Bost F., Luo C., Birrer M. J., Gjerset R., fects caused by treatment with high doses of cisplatin. Mercola D., J. Biol. Chem., 272, 14041—14044 (1997). 34) Sanchez-Perez I., Perona R., FEBS Lett., 18, 151—158 (1999). REFERENCES 35) Woessmann W., Chen X., Borkhardt A., Cancer Chemother. Pharma- col., 50, 397—404 (2002). 1) Lee W. M., Semin. Liver Dis., 23, 217—226 (2003). 36) Persons D. L., Yazlovitskaya E. M., Cui W., Pelling J. C., Cancer Res., 2) Park B. K., Kitteringham N. R., Maggs J. L., Pirmohamed M., 5, 1007—1014 (1999). Williams D. P., Annu. Rev. Pharmacol. Toxicol., 45, 177—202 (2005). 37) Chang L., Karin M., Nature (London), 410, 37—40 (2001). 3) Calvert A. H., Horwich A., Newlands E. S., Begent R., Rustin G. J., 38) Devary Y., Gottlieb R. A., Lau L., Karin M., Mol. Cell Biol., 11, Kaye S. B., Harris A. L., Williams C. J., Slevin M. L., Lancet, 3, 2804—2811 (1991). 577—578 (1988). 39) Bossy-Wetzel E., Bakiri L., Yaniv M., EMBO J., 16, 1695—1709 4) Loehrer P. J., Einhorn L. H., Ann. Intern. Med., 100, 704—713 (1984). (1997). 5) Waszkiewicz K., Postepy Hig. Med. Dosw., 55, 387—406 (2001). 40) Hilberg F., Wagner E. F., Oncogene, 7, 2371—2380 (1992). 6) Arany I., Safirstein R. L., Semin. Nephrol., 23, 460—464 (2003). 41) Ryseck R. P., Hirai S. I., Yaniv M., Bravo R., Nature (London), 334, 7) Cetin R., Devrim E., Kilicoglu B., Avci A., Candir O., Durak I., J. 535—537 (1988). Appl. Toxicol., 26, 42—46 (2006). 42) Shaulian E., Karin M., Nat. Cell Biol., 4, 131—136 (2002). 8) Price P. M., Safirstein R. L., Megyesi J., Am. J. Physiol. Renal Physiol., 43) Kasibhatla S., Brunner T., Genestier L., Echevarr F., Mahaboubi A., 286, 378—384 (2003). Green D., Mol. Cell, 1, 543—551 (1998). 9) Cavalli F., Tschopp L., Sonntag R. W., Zimmermann A., Cancer Treat. 44) Kolbus A., Herr I., Schreiber M., Debatin K. M., Wagner E. F., Angel Rep., 62, 2125—2126 (1978). P., Mol. Cell. Biol., 20, 575—582 (2000). 10) Cersosimo R. J., Ann. Pharmacother., 27, 438—441 (1993). 45) Minden A., Karin M., Biochim. Biophys. Acta, 1333, 85—104 (1997). 11) Pollera C. F., Ameglio F., Nardi M., Vitelli G., Marolla P., J. Clin. 46) Smeal T., Angel P., Meek J., Karin M., Genes Dev., 3, 2091—2100 Oncol., 5, 318—319 (1987). (1989). 12) Olajide O. A., Ajayi F. F., Ekhelar A. I., Awe S. O., Makinde J. M., 47) Gunawan B. K., Liu Z. X., Han D., Hanawa N., Gaarde W. A., Alada A. R., Phytother. Res., 13, 344—345 (1999). Kaplowitz N., Gastroenterology, 131, 165—178 (2006). 13) Ozaki Y., Soedigdo S., Wattimena Y. R., Suganda A. G., Jpn. J. Phar- 48) Mandlekar S., Kong A. N., Apoptosis, 6, 469—477 (2001). macol., 49, 155—163 (1989). 49) Menor C., Fernández-Moreno M. D., Fueyo J. A., Escribano O., 14) Sonavane G. S., Sarveiya V. P., Kasture V. S., Kasture S. B., Pharma- Olleros T., Arriaza E., Cara C., Lorusso M., Di Paola M., Román I. D., col. Biochem. Behav., 71, 239—244 (2002). Guijarro L. G., J. Pharmacol. Exp. Ther., 311, 668—676 (2004). 15) Lee J. Y., Han Y. B., Woo W. S., Shin K. H., Kor. J. Pharmacog., 21, 50) Kim, S. H., Hong K. O., Hwang J. K., Park K. K., Food Chem. 270—273 (1990). Toxicol., 43, 117—122 (2005). 16) Shin K. H., Woo W. S., Kor. J. Pharmacog., 17, 91—99 (1986).