The Anti-Tumor Effects of Alkaloids from the Seeds of Strychnos Nux-Vomica

Journal of Ethnopharmacology 106 (2006) 179–186

The anti-tumor effects of alkaloids from the seeds of Strychnos nux-vomica on HepG2 cells and its possible mechanism Xu-Kun Deng a,1,WuYinb,1, Wei-Dong Li a, Fang-Zhou Yin a, Xiao-Yu Lu a, Xiao-Chun Zhang a, Zi-Chun Hua b, Bao-Chang Cai a,∗ a College of Pharmacy, Nanjing University of Chinese Medicine, Nanjing 210029, China b State Key Lab of Pharmaceutic Biotechnology, Nanjing University, Nanjing 210093, China Received 6 April 2005; received in revised form 12 December 2005; accepted 15 December 2005 Available online 26 January 2006

Abstract To screen the anti-tumor effects of the four alkaloids: brucine, strychnine, brucine N-oxide and isostrychnine from the seed of Strychnos nux- vomica, MTT assay was used to examine the growth inhibitory effects of these alkaloids on human hepatoma cell line (HepG2). Brucine, strychnine and isostrychnine revealed significant inhibitory effects against HepG2 cell proliferation, whereas brucine N-oxide didn’t have such an effect. In addition, brucine caused HepG2 cell shrinkage, membrane blebbing, apoptotic body formation, all of which are typical characteristics of apoptotic programmed cell death. The results of flow cytometric analysis demonstrated that brucine caused dose-dependent apoptosis of HepG2 cells through cell cycle arrest at G0/G1 phase, thus preventing cells entering S or G2/M phase. Immunoblot results revealed that brucine significantly decreased the protein expression level of cyclooxygenase-2, whereas increased the expression caspase-3 as well as the caspase-3-like protease activity in HepG2 cells, suggesting the involvement of cyclooxygenase-2 and caspase-3 in the pro-apoptotic effects exerted by brucine. Therefore, this paper indicate that the major alkaloids present in the seed of Strychnos nux-vomica are effective against HepG2 cells proliferation, among which brucine proceed HepG2 cells death via apoptosis, probably through the participation of caspase-3 and cyclooxygenase-2. © 2005 Elsevier Ireland Ltd. All rights reserved.

Keywords: Strychnos nux-vomica; Brucine; HepG2; Apoptosis; Cyclooxygenase-2; Caspase-3

1. Introduction tablet in liver cancer therapy. In addition to its anti-tumor effects, nux vomica is also claimed to have an ability of improving cir- Hepatocellular carcinoma is one of the most common malig- culatory system and relieving pains in patients with rheumatic nant tumors worldwide. In last decades, surgical resection, the disorders (Guizhi, 1996). However, the mechanism underly- commonly used technique is frequently challenged in view of ing the anti-tumor property of the nux vomica remains largely metastasis and other pathological changes. Therefore, the devel- unknown. opment of new agents for hepatocellular cancer is important to Alkaloids are the main bioactive chemicals in nux vomica reduce the mortality caused by this disease. (Bisset and Phillipson, 1971), responsible for the pharmacolog- Nux vomica, the dried seed of Strychnos nux-vomica L. ical and toxic effects exerted by nux vomica to a great extent. In (Loganiaceae), has been effectively used in Chinese folk our previous study, 16 alkaloids have been separated and iden- medicine for the treatment of liver cancer and associated patho- tiﬁed from the crude nux vomica, among which strychnine and logical abnormalities for a long history. As a major ingredient, it brucine take up 80% (Cai et al., 1994). Based on the principles has been frequently prescribed in some classic herbal remedies of Chinese medicine, nux vomica is commonly prescribed in such as “Ping-xiao” capsule, “Ci-dan” capsule, “Ma-qian-zi” clinical practice after sand processing, the purpose of which is not only to clean its ﬁne hairs which causes throat irritation (Wu et al., 1994), but also reduce its toxic side effects because after processing, the intrinsic alkaloids such as brucine and strychnine ∗ Corresponding author. Tel.: +86 25 86798281; fax: +86 25 86798281. E-mail address: [email protected] (B.-C. Cai). transform into their isoforms or nitrogen oxidative derivatives 1 These authors contribute equally to this work. with less toxicity (Fig. 1)(Cai et al., 1990, 1993). However, the

Fig. 1. Chemical structures of brucine, brucine N-oxide, strychnine, isostrychnine, major present alkaloids in nux vomica and the transformations between them after hot sand processing. (1) Strychnine transformed into its isoform, isostrychnine after sand processing. (2) Brucine transformed into its nitrogen derivatives, brucine N-oxide after sand processing. alterations in chemical structures also result in different phar- Sigma Chemical Co. (St. Louis, MO, USA); brucine and brucine macological effects possessed by these alkaloids. For example, N-oxide were supplied by Carl Roth GmbH+ Co. (German); brucine, in crude nux vomica, was claimed to be a morphine-like isostrychnine was obtained from nux vomica based on the pre- analgesic substance, whereas its derivatives brucine N-oxide, in viously described method (Cai et al., 1994). In brief, after sand processed nux vomica, act as a more NSAIDs-like com- extraction from nux vomica, isostrychnine was confirmed by pound (Yin et al., 2003a). comparing the physical and chemical properties such as melting In our previous study, the total alkaloid fractions from nux point and thin layer chromatography with standard compounds vomica were shown to remarkably suppress the HeLa and K562 and 1H NMR spectra. Its purity was more than 99%. The seeds cells growth (Cai et al., 1995, 1998), the results of which pro- of Strychnos nux-vomica L. were collected and identified by vided the possibility that the traditional application of nux vom- professor Bao-Chang Cai and De-Kang Wu from Nanjing Uni- ica in the therapy of liver cancer is also due to the direct cytotoxic versity of Chinese Medicine. A voucher sample (Wu 200052) effects on liver tumor cells elicited by these alkaloids. Therefore, was deposited in the College of Pharmacy, Nanjing University in order to test this hypothesis and identify the alkaloid which of Chinese Medicine. All water soluble compounds were dis- is mainly responsible for the anti-tumor effect of nux vomica solved in RPMI 1640 medium (Gibco, USA), whereas insoluble as well as its mechanisms, we planned to screen the cytotoxic chemicals were dissolved into DMSO, the final concentration effects of strychnine, brucine, isostrychnine and brucine N-oxide of DMSO in each sample were less than 0.1%. Adjustment as described above on HepG2 cells, and choose a representative of the solutions to pH 7.2–7.4 was made with HCl (0.5 M) or compound to further investigate its anti-tumor mechanisms. In NaOH (0.5 M) if necessary. All solutions were passed through these four alkaloids, strychnine and brucine were selected to rep- a 0.22 ␮m filter (GVMP 01230, Millipore) and stored at 4 ◦C resent the alkaloids stemmed from crude nux vomica, whereas until use. isostrychnine and brucine N-oxide stood for the alkaloids from processed nux vomica. The results of this investigation might 2.2. Cell culture provide a scientific explanation for the traditional application of this herbal medicine in liver cancer therapy. Human hepatoma cell lines (HepG2) was purchased from cell bank of Shanghai Institute of Cell Biology (Shanghai, 2. Materials and methods China), maintained in RPMI 1640 culture medium plus 10% calf serum and 100 U/ml penicillin, 75 U/ml streptomycin, in ◦ 2.1. Materials a37 C incubator supplied with 95% room air and 5% CO2. After 60–80% confluency, the cells were trypsinized with 0.25% Strychnine, Hoechst 33258, 3-(4,5-dimethylthiazol-2-yl)- trypsin (AMRESCO, dissolved in PBS, pH 7.4), counted and 2,5-diphenylterazolium bromide (MTT) were purchased from placed down at needed density for treatment. X.-K. Deng et al. / Journal of Ethnopharmacology 106 (2006) 179–186 181

2.3. MTT-colorimetric assay confluent cells were lysed and collected, resolved by a 12% SDS- PAGE gel, the protein was then electrophoretically blotted onto Assessment of in vitro neoplastic activity was determined polyvinylidene fluoride (PVDF) membranes. The membranes based on the previously described method (Fatope et al., 1996). were first hybridized with primary antibodies and then with Briefly, the HepG2 cells in exponential growth were placed down horseradish peroxidase-conjugated anti-mouse or anti-rabbit at a final concentration of 6 × 104 cells/ml in 96-well plate. IgG secondary antibody (Sigma, St. Louis, MO, USA). The After 40% confluency, the cells were treated with strychnine, antibody against the human cyclooxygenase-2 was purchased brucine, brucine N-oxide or isostrychnine at varying concen- from Santa Cruz Biotech (Santa Cruz, CA, USA), antibody trations, 100 ␮l MTT (1 mg/ml) was then added at 24, 48, 72 h against glyceraldehyde-3-phosphate dehydrogenase (GAPDH) after treatments. For MTT assay, the supernatant was discarded were purchased from Upstate Biotechnology (Lake Placid, NY, and 200 ␮l DMSO was added, the 96 wells plate was vibrated USA). The immune complexes were detected using chemilumi- on micro-vibrator for additional 10 min, the optical density of nescence (ECL) system (Amersham Pharmacia Biotech, Buck- each well was measured at λ492nm by Enzyme-immunoassay inghamshire, UK). instrument (Nanjing, China). The growth inhibitory ratio of each alkaloid was calculated based on the following formula: (average O.D. of control group − average O.D. of treated group) Inhibitory ratio of growth (%) = × 100 average O.D. of control group

2.4. Apoptosis analysis by Hoechst 33258 staining 2.7. Measurement of caspase-3-like activity The Hoechst 33258 staining was carried out as previously described (Ramonede and Tomas, 2002). Briefly, the HepG2 The activity of caspase-3-like proteases was mea- cells in exponential growth were placed down at a final con- sured as increases in hydrolysis of fluorogenic tetrapeptide centration of 1 × 105/well in 12-well culture plate. A cover slip substrate, Ac-DEVD-7-amino-4-methylcoumarin (Ac-DEVD- was placed on the bottom of each well to allow cells to grow on AMC), according to the manufacturer’s instructions (BIOMOL its surface as a monolayer. After 40–60% confluency, the cells Research Laboratories, Inc., Plymouth Meeting, PA, USA). were exposed to brucine 0.25 mM for 36 h. For Hoechst 33258 Because caspase 7 also cleaves the substrate, the activity staining, the cells were fixed with ice-cold methanol, acetic acid obtained here represents as caspase-3-like proteases activity. (3:1) at room temperature for 5 min after washing twice with Briefly, the HepG2 cells were exposed to brucine (0.125, 0.25, ice-cold PBS, the re-washed with ice-cold PBS buffer twice, 0.5, 1.0 mM) for 36 h, after treatment, the cells were lysed, the supernatant was taken for measurement of hydrolysis of Ac- loaded with Hoechst 33258 for additional 20 min, the changes ◦ in the nuclei of cells after Hoechst 33258 staining were observed DEVD-AMC as a function of time at 22 C. under a fluorescence microscope (Olympus, BX-60, Japan) in less than 15 min. 2.8. Statistical analysis

2.5. DNA content and cell cycle analyzed by flow cytometry The results were expressed as mean ± S.D. The statistical analysis involving two groups was performed by means of Stu- The DNA content and the cell cycle of HepG2 cells dent’s t-test, whereas analysis of variance (ANOVA) followed by were determined by flow cytometry based on previously Dunnett’s multiple comparison test were used in order to com- described method (Yin et al., 2003). Briefly, the HepG2 cells pare more than two groups. All data were processed with SPSS were placed down at a density of 1 × 106 cells/well in a 6- software. P < 0.05 was considered significant, and P < 0.01 was well plate. After 40% confluency, the cells were exposed to considered more significant. Regression analysis was used to brucine (0.25, 0.5, 1.0 mM) for additional 36 h. Cells were calculate IC50 of tested chemicals. collected, washed twice with ice-cold PBS buffer (pH 7.4), fixed with 70% alcohol and stained with propidium iodide 3. Results (PI) (1 mg/ml) in the presence of 1% RNAase A atleast for 30 min before analysis by flow cytometry (Becton Dickinson, 3.1. Growth inhibitory effects of strychnine, brucine, USA). The data were analyzed with MODFIT and CellQuest brucine N-oxide, isostrychnine on HepG2 cells by MTT software. assay

2.6. Immunoblot analysis MTT assay was used as an indirect measure to determine the viability of HepG2 exposed to the alkaloids. As shown The immunoblot analysis of caspase-3 and cyclooxygenase- in Fig. 2(1)–(3), both brucine and isostrychnine caused the 2 were performed as previously described (Yin et al., 2003b). cell death in a concentration, time-dependent manner. Brucine Brieﬂy, the HepG2 cells were placed down at a ﬁnal density (1 mM) showed the strongest cytotoxic effect on HepG2 at 72 h of 8 × 105 cells/well in a 12-well plate. After treatments, the after treatment (96.3%), whereas the maximal growth inhibitory 182 X.-K. Deng et al. / Journal of Ethnopharmacology 106 (2006) 179–186

Fig. 2. The growth inhibitory effects of strychnine, isostrychnine, brucine, brucine N-oxide on HepG2 cells at different time after treatment. The HepG2 (6 × 104 cells/ml) was incubated with the indicated four alkaloids at 0.125, 0.25, 0.5, 1 mM for 24 h (panel 1), 48 h (panel 2) and 72 h (panel 3). The cell viability was measured in three plates in triplicate using MTT assay. *P < 0.05, **P < 0.01 as compared to medium alone control. rate for isostrychnine was 72.7%. As an analogue to isostrych- more effective against HepG2 cell proliferation than the latter. nine, strychnine demonstrated the milder cytotoxic effect as The IC50 value of brucine N-oxide at any given time exceeded compared to that of isostrychnine. Only after incubation with 1.0, further revealed its ineffectiveness on the HepG2 cells pro- strychnine for 72 h, the HepG2 cells died in a dose-dependent liferation. manner. However, its maximal cytotoxic effect (89%) was very close to that of brucine, even slightly larger than isostrychnine. 3.2. Morphological changes of HepG2 cells induced by In contrast to strychnine, brucine N-oxide didn’t present any brucine under ﬂuorescent microscope cytotoxic effect at any given time or concentrations. The cytotoxic effects of the four alkaloids were also revealed from the Hoechst staining was applied to investigate whether HepG2 results in Table 1, aside from brucine N-oxide and isostrychnine underwent cell death via apoptosis or necrosis. As shown at 24 h after treatment, the IC50 of the cytotoxic effects of strych- in Fig. 3, after Hoechst staining, the HepG2 cells exposed nine, isostrychnine or brucine at 72 h was below 1.0, the value of to brucine (0.25 mM) revealed marked nuclear condensation, which was considered to be effective enough to inhibit the cell membrane blebbing, nuclear fragmentation and apoptotic bod- growth. Brucine had a much smaller IC50 value as compared to ies (as illustrated by arrows), all of which are the characteristics that of strychnine or isostrychnine, suggesting the former was of apoptotic programmed cell death. However, the control cells didn’t display any of the given characteristics.

Table 1 3.3. Cell cycle analysis by ﬂow cytometry IC50 (mM) effects of strychnine, brucine, brucine N-oxide, isostrychnine on the HepG2 cell growth at different time after treatment The results presented in Fig. 4(1)–(4) showed that after Strychnine Brucine Brucine N-oxide Isostrychnine treatment for 36 h, brucine dose-dependently increased the sub 24 h 1.00 0.65 1.00 >1.00 G0/G1 population, which indicated the increased cell apopto- 48 h 1.00 0.32 1.00 0.98 sis. Also from the results of cell cycle distribution in Table 2, 72 h 0.52 0.10 1.00 0.61 HepG2 cells were arrested at G0/G1 phase, thus were prevented X.-K. Deng et al. / Journal of Ethnopharmacology 106 (2006) 179–186 183

Fig. 3. Morphological alterations of HepG2 cells following exposure to brucine (0.25 mM) for 36 h. The HepG2 cells were grown on the cover slip as a monolayer, after treatment, the control cells (panel 1), brucine-treated cells (panel 2) were loaded with Hoechst 33258 and observed under the ﬂuorescence microscope. Arrows A (panel 2): nuclear shrinking; B: membrane blebbing; C: apoptosis body. Each experiment was performed in triplicate (magniﬁcation 100×).

Fig. 4. Flow cytometric analysis of HepG2 cells following exposed to brucine. Cells were treated with no brucine (panel 1), brucine 0.125 mM (panel 2), brucine 0.25 mM (panel 3), brucine 0.5 mM (panel 4) for 36 h, stained with PI and analyzed on a FACScalibur ﬂow cytometer. The apoptotic rate and cell cycle distribution of each group were stated in Table 2. Each experiment was performed in triplicate. 184 X.-K. Deng et al. / Journal of Ethnopharmacology 106 (2006) 179–186

Table 2 Cell cycle distributions of HepG2 after treatment with brucine for 36 h Group Apoptotic Cell cycle distributions of HepG2 (%) rate (%) G0/G1 G2/M S Control 4.52 47.55 12.02 40.43 Brucine 0.125 mM 28.15 53.61 6.89 39.50 Brucine 0.25 mM 36.76 58.76 7.15 34.10 Brucine 0.5 mM 48.49 60.80 11.05 28.15 from entering S phase since the population in G2/M or S phase declined, whereas in G0/G1 phase increased.

3.4. Effects of brucine on the expression of caspase-3 and cyclooxygenase-2 protein

To further explore the apoptotic mechanism of HepG2 cells Fig. 6. Effect of brucine on the caspase-3-like activity. The hydrolysis of Ac- induced by brucine, western blot was performed to examine DEVD-AMC was considered as the indicator for the caspase-3-like activity. Cells were treated with brucine at 0.125, 0.25, 0.5, 0.1 mM for 36 h. After treat- the protein expression level of caspase-3 and cyclooxygenase- ment, the cells were lysed, the caspase-3-like activity of the supernatant was 2 in brucine-treated cells. As shown in Fig. 5, brucine measured. n =3,**P < 0.01, compared with control. induced concentration-dependent increase of caspase-3-protein abundance associated with the corresponding decrease of activity. The strongest hydrolysis effect was exerted by brucine at cyclooxygenase-2 expression. Sodium diclofenac (100 ␮g/ml), 0.5 mM after incubation for 36 h (72.41%). The highest concen- a commonly used NSAIDs, served as a control, showed the sim- tration of brucine (1.0 mM), however, decreased the hydrolysis ilar inhibitory effect on the expression of cyclooxygenase-2 as of Ac-DEVD-AMC as compared to other lower concentrations that of brucine, but didn’t reveal any signiﬁcant effect on the (63.76%), perhaps due to the larger cell death rate. level of caspase-3.

3.5. Effect of brucine on the caspase-3-like activity 4. Discussion and conclusions

Since brucine induced the over-expression of caspase-3 In the present study, we investigated the cytotoxic effects in HepG2 cells, Ac-DEVE-AMC, the normally used peptide of four alkaloids originated from crude or processed nux vom- substrate to measure the caspase-3-like protease activity, was ica on HepG2 cells and the possible mechanisms. We found employed to determine the caspase-3-like activity in HepG2 that brucine, among the selected alkaloids, exhibited a strongest cells after treatment with brucine, the results in Fig. 6 demon- inhibitory effect on HepG2 cells proliferation. In contrast to strated that brucine dose-dependently increased the hydrolysis brucine, brucine N-oxide didn’t reveal any anti-proliferation of Ac-DEVD-AMC, an indicator of the caspase-3-like protease activity. Strychnine, isostrychnine also demonstrated significant inhibitory effects on HepG2 cells growth, although their activities were less potent than that of brucine. Furthermore, brucine was found to cause HepG2 cells via caspase-3 and cyclooxygenase-2-mediated cells apoptosis. In our previous study, brucine N-oxide was found to pos- sess a more potent inhibitory effect on the ADP-induced platelet aggregation than that of brucine (Jianyin et al., 1998), also the anti-inflammatory effect of brucine N-oxide was stronger than that of brucine in animal test (Yin et al., 2003a). However, brucine N-oxide in this study didn’t show any inhibitory effect on HepG2 cells proliferation. This phenomenon implied that the pharmacological effects of brucine and brucine N-oxide were strikingly different despite that only minor variations existed in their chemical structures. Interestingly, this large difference Fig. 5. Immunoblot analysis of the protein abundance of caspase-3 and in structure–activity relationships of brucine didn’t applied to cyclooxygenase-2 of HepG2 cells exposed to brucine. The HepG2 cells were strychnine, since both strychnine and isostrychnine demon- treated with 0.125 mM brucine (B), 0.25 mM (C), 0.5 mM (D) for 36 h, Sodium strated significant inhibitory effects on HepG2 cells prolifera- diclofenac (100 ␮g/ml) (E) served as a positive control, cells with culture medium alone (A) were included as a negative control. GAPDH was used as tion, although their effects were less potent than that of brucine. an internal standard to normalize loadings. Each experiment was performed in Besides, based on the above results, it seemed that the crude nux triplicate. vomica was more advantageous over the processed one in treat- X.-K. Deng et al. / Journal of Ethnopharmacology 106 (2006) 179–186 185 ing liver cancer. However, in our recent study, we found that the adjuvant-induced arthritis, the action mode of which is similar to processed nux vomica was more effective against HepG2 cells the common NSAIDs (Yin et al., 2003a). Given the above facts, proliferation than that of crude nux vomica (unpublished data). brucine could be a promising NSAID, which was effective in the In detail, the IC50 effect of the total alkaloids from crude nux treatment of liver cancer associated with pain or inflammation. vomica on HepG2 cells proliferation after treatment for 72 h was Finally, we have to confess that the concentration brucine 53.1 mg/l, whereas the alkaloids from processed nux vomica was used in this study was a little bit higher than its physiological 37.8 mg/l. This data suggested that brucine may not fully rep- concentration, which may not be directly translated into clinical resent the anti-proliferation effect of alkaloids in nux vomica, practices. However, the effect of brucine, atleast in this study some other alkaloids such as isobrucine may also make a con- has been proven to be specific because brucine at 0.5 mM or tribution to the anti-proliferation effect of nux vomica, although higher didn’t cause significant human embryonic kidney cells they are present in nux vomica in small quantities. In addition, (HEK293) or other mammalian cells death (data not shown). we couldn’t formally exclude other mechanisms such as anti- Taken together, we have successfully established an in vitro angiogenic or immunostimulant effects were also involved in model to evaluate the possible mechanism of alkaloids in nux the anti-tumor property of alkaloids in nux vomica. Thus, more vomica in the therapy of liver cancer. The final goal we are will- experiments needs to be done to further identify the mechanisms ing to achieve is to find a more effective anti-tumor drug by of nux vomica in live cancer therapy. In fact, there still existed properly modifying the structure of brucine, since many alka- another advantage for processed nux vomica in treating liver can- loids such as harringtonine, vincaleukoblastine and vincristine cer associated with inflammation, because the abundant brucine which have strong anti-tumor effects share the indole ring in N-oxide in processed nux vomica may inhibit the inflammatory their chemical structures with brucine. Thus, the experimental reactions in tumor-bearing patients as previously described (Yin model established in this paper will surely accelerate the under- et al., 2003a). standing of the structure–activity relationships in brucine and During the past decades, the killing of tumors through the pave the way for discovering new drug. induction of apoptosis has been recognized as a novel strategy for the identification of anticancer drugs (Panchal, 1998; Smets, Acknowledgements 1994; Watson, 1995). The apoptosis itself also plays an important role in the development of various diseases including cancer This work was supported by National Nature Science Fund of (Fisher, 1994; McConkey et al., 1996). In the present study, China (90209052, 30470644), PhD Candidate Fund of Jiangsu brucine exhibited a caspase-3-dependent HepG2 cells apopto- Province (2005098). The authors express thanks for the assis- sis since the protein abundance of caspase-3 and caspase-3-like tance from vice professor Wang Tian-Shan and Di Liu-Qing protease activity in HepG2 cells were dramatically up-regulated from the College of Pharmacy, Nanjing University of Chinese by brucine. Also, brucine was shown to cause HepG2 cells Medicine for the generation of the original research proposal, cycle arrest at G0/G1 phase, thus preventing cells from enter- also critical comments on the direction of this project and ing S or G2/M phase and finally apoptosis occurred. Therefore, great technical assistance from Xu Dong-Qing, Wang Ming- whether the hindrance of HepG2 cells cycle progression caused Yan, Xiang Xiao-Ren, Sheng Yu-Qing. by brucine is related to caspase-3 needs more experiments to clarify. In the present study, brucine was also shown to reduce the References expression of cyclooxygenase-2 in HepG2 cells. Cyclooxyge- nase is a rate-limiting enzyme in prostanoids (PGs) biosynthesis. Bisset, N.G., Phillipson, J.D., 1971. The tertiary alkaloids of some Asian species of strychnos. Journal of Pharmaceutical and Pharmacology 23, It consists of two isozymes, the constitutive cyclooxygenase- 244S. 1 and the inducible cyclooxygenase-2. The cyclooxygenase-2 Cai, B.C., Hattori, M., Namba, T., 1990. Processing of nux vomica. II. gene has been characterized as an immediate early gene and Changes in alkaloid composition of the seeds of Strychnos nux-vomica associated with inflammation, pain and cellular proliferation L. on traditional drug-processing. Chemical and Pharmaceutical Bulletin as well as apoptosis. Recently, lines of evidences suggest that 38, 1295–1298. Cai, B.C., Yang, W.X., Zhu, W.Y., Lu, J.C., Ye, D.J., 1993. Effect of process- the over-expression of cyclooxygenase-2 might be one of the ing on the extraction of alkaloids from Strychnos nux-vomica L. Journal leading factors in hepatic carcinogenesis (Koga et al., 1999). of Chinese Herbal Medicine 18, 23, 62. Non-steroid anti-inflammatory drugs (NSAIDs) are commonly Cai, B.C., He, Y.W., Ding, H.F., Ma, C., Wu, H., 1994. Determination of used drugs to specially inhibit the cyclooxygenase-2 activity. alkaloids and comparison of the acute toxicity of differently processed NSAIDs can inhibit the tumor cell proliferation, induce apop- products of the seeds of Strychnos nux-vomica L. Journal of Chinese Herbal Medicine 19, 598–600, 638. tosis through down-regulation of bcl-2 expression and levels Cai, B.C., Chen, L., Namba, T., 1995. Processing of nux vomica.IV.A of prostanoids (Li et al., 2002; Sheng et al., 1997). A case in comparison of nine alkaloids from processed seeds of Strychnos nux- point is nimesulide, which suppress tumor growth and induce vomica L. on tumor cell lines. Natural Medicines 49, 39–42. apoptosis in implanted hepatoma mice by inhibiting the expres- Cai, B.C., Wang, T.S., Kurokawa, M., Shiraki, K., Hattori, M., 1998. Cyto- toxicities of alkaloids from processed and unprocessed seeds of Strychnos sion of cyclooxygenase-2 and prostaglandin E2 (Li et al., 2003; nux-vomica L. Acta Pharmacologica Sinica 19, 425–428. Tsujii and DuBois, 1995). In our previous study, brucine was Fatope, M.O., Zeng, L., Ohayaga, J.E., Shi, G., Mclaughlin, J.L., 1996. Selec- shown to relieve pain, reduce the release of inflammatory factors tively cytotoxic diterpenes from Euphorbia poisonii. Journal of Medicinal such as prostaglandin E2, thromboxane B2 and 6-keto-PGF1a in Chemistry 39, 1005–1008. 186 X.-K. Deng et al. / Journal of Ethnopharmacology 106 (2006) 179–186

Fisher, D.E., 1994. Apoptosis in cancer therapy: crossing the threshold. Cell Ramonede, B.M., Tomas, R.P., 2002. Activation of protein kinase C for 78, 539–542. protection of cells against apoptosis induced by the immunosuppressor Guizhi, M., 1996. The long period clinical observation of the effect of Strych- prodigiosin. Biochemical Pharmacology 63, 463–469. nos nux-vomica on Kaschin-Beck’s disease. Chinese Journal of Regional Sheng, H., Shao, J., Kirkland, S., 1997. Inhibition of human colon cancer cell Diseases. Prevention and Therapy 11, 120–124. growth by selective inhibition of cyclooxygenase-2. Journal of Clinical Jianyin, Z., Huimin, B., Chen, M., 1998. Study on the effect of brucine and Investigation 99, 2254–2259. brucine N-oxide on platelet aggregation and blood clot formation. Jiangsu Smets, L.A., 1994. Programmed cell death (apoptosis) and response to anti- Journal of Chinese Traditional Medicine 19, 41–42. cancer drugs. Anticancer Drugs 5, 3–9. Koga, H., Sakisaka, S., Ohishi, M., 1999. Expression of cyclooxygenase-2 Tsujii, M., DuBois, R.N., 1995. Alterations in cellular adhesion and apoptosis in human hepatocellular carcinoma: relevance to tumor dedifferentiation. in epithelial cells over expression of prostaglandin endoperoxide synthase Hepatology 29, 688–690. 2. Cell 83, 493–501. Li, H.L., Zhang, H.W., Chen, D.D., Zhong, L., Ren, X.D., Tu, R., 2002. Watson, A.J., 1995. Review article: manipulation of cell death—the devel- JTE-522, a selective COX-2 inhibitor, inhibits cell proliferation and opment of novel strategies for the treatment of gastrointestinal disease. induces apoptosis in RL95-2 cells. Acta Pharmacologica Sinica 23, 631– Alimentary Pharmacology and Therapeutics 9, 215–226. 637. Wu, H., Cai, B.C., Zheng, P.X., Zhao, C.Z., Yuan, Z.R., 1994. Effect of Li, X.H., Li, J.J., Zhang, H.W., 2003. Nimesulide inhibits tumor growth processing on the alkaloids in Strychnos nux-vomica L. Journal of Chinese in mice implanted hepatoma: overexpression of Bax over Bcl-2. Acta Herbal Medicine 19, 277–279. Pharmacological Sinica 24, 1045–1050. Yin, W., Wang, T.S., Yin, F.Z., Cai, B.C., 2003a. Analgesic and anti-inﬂam- McConkey, D.J., Zhivotovsky, B., Orrenius, S., 1996. Apoptosis molecular matory properties of brucine and brucine N-oxide extracted from seeds mechanisms and biomedical implications. Molecular Aspects of Medicine of Strychnox nux-vomica. Journal of Ethnopharmacology 88, 205–214. 17, 1–110. Yin, W., Zhou, X.M., Cai, B.C., 2003b. Mechanism of cytotoxicity of human Panchal, R.G., 1998. Novel therapeutic strategies to selectively kill cancer embryonic kidney cells induced by gliotoxin. Chinese Journal of Phar- cells. Biochemical Pharmacology 55, 247–252. macology and Toxicology 17, 321–327.