Centella Asiatica Extract Induces Cell Cycle Arrest in Caco-2 Human Colon Cancer Cells

Chiang Mai Med Bull 2005;44(1):21-28.

Original article

CENTELLA ASIATICA EXTRACT INDUCES CELL CYCLE ARREST IN CACO-2 HUMAN COLON CANCER CELLS

Piyawan Bunpo,1,2 B.Sc., Keiko Kataoka,2 Ph.D., Hideki Arimochi,2 Ph.D., Haruyuki Nakayama,2 M.Sc., Tomomi Kuwahara,2 Ph.D., Yoshinari Ohnishi,2 Ph.D., Usanee Vinitketkumnuen,1 Ph.D.

1Department of Biochemistry, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand, 2Department of Molecular Bacteriology, Graduate School of Medicine, The University of Tokushima, Tokushima, Japan 770-8503

Abstract Centella asiatica is traditionally a medicinal plant frequently employed in the practice of Thai folk medicine. This study examined the anti-tumor activity of the crude water extract of C. asiatica using human colon adenocarcinoma-derived Caco-2 cells. C. asiatica extract reduced the proliferation rate of Caco-2 cells significantly in a concentration- and time-dependent manner. The mechanism of cancer cell growth inhibition

was shown to occur via cell cycle arrest. The extract induced S and G2-M arrest in Caco-2 cells accompanied apoptosis induction. The extract also increased the accumulation of cyclin B1 protein in the cells. These findings indicate that C. asiatica extract inhibited cell proliferation of Caco-2 cells through modification of the cell cycle events and this cell cycle arrest is associated, at least in part, with increased accumulation of cyclin B1 protein. Chiang Mai Med Bull 2005;44(1):21-28.

Keywords: centella asiatica, colon adenocarcinoma cell lines, cell proliferation, cell cycle arrest, cyclin B1

The research for bioactive natural has shown that some medicinal plants products and their analogues as chemo- exhibit inhibitory effects on the forma- therapeutic agents continues in preclini- tion of azoxymethane (AOM)-induced cal development.(1) Various phytochemi- aberrant crypt foci (ACF).(3-6) Centella cals originating from medicinal plants asiatica Linn. is a medicinal plant tradi- have been shown to have anti-tumor tionally used in Asia. This plant posseses activities, and some of them are currently many kinds of biological activities such used in clinical fields.(2) Our laboratory as the elevation of antioxidant level in

Address requests for reprints: Usanee Vinitketkumnuen, Ph.D. Department of Biochemistry, Faculty of Medicine, Chiang Mai University, Chiang Mai 50200, Thailand Received 17 September 2004, and in revised form 11 November 2004. 22 Bunpo P, et al. wound healing(7) and anti-herpes simplex Materials and methods virus activity.(8) Crude extracts of this Preparation of plant extract plant have been shown to inhibit the proli- C. asiatica was purchased from local feration of transformed cell lines and markets in Chiang Mai, Thailand. The ascites tumor cells and retard the de- fresh plant (100 g) was washed with water, velopment of solid and ascites tumors.(9) cut into small pieces and extracted with Moreover, our previous report found that 200 mL of cold distilled water for 4 hr. C. asiatica extract inhibited the forma- The extract was centrifuged at 4,000xg for tion of AOM-induced ACF in rat colons 15 min before the supernatant was filtered and had a chemopreventive effect on through paper (Whatman No.1). The fil- colon tumorigenesis.(10) Principle active trate was evaporated in a rotatory evapo- components of C. asiatica are pentacy- rator under reduced pressure at 45-50 oC clic triterpenes.(11) prior to lyophilization. The yield of water Cellular homeostasis critically depends extract from the fresh plant was 2.5%. on a balance of cell proliferation, dif- ferentiation and death.(12) Since uncon- Cell line and culture condition trolled cell proliferation has essential Caco-2 cells (RCB0988, Riken Cell roles in carcinogenesis, regulation of cell Bank, Tsukuba Science City, Japan) were proliferation is important for cancer pre- cultured in MEM (ICN Biomedical, vention.(13) Some phytochemicals have Ohio, USA) containing 20% heat-inacti- been reported to show anti-tumor activi- vated fetal bovine serum (FBS, GIBCO- ty by apoptosis induction and/or cell BRL, Grand Island, NY), 100 units/mL cycle arrest.(14-15) Cell cycle is regulated penicillin, 100 µg/mL streptomycin sulfate by two processes: oscillating changes in and 2 mM glutamine. Cells were grown the activity of the cell cycle machinery at 37 oC in a fully humidified atmos- and specific proteolysis of the cell cycle phere containing 5% CO2. The number regulators.(16) Cyclin-dependent kinase of viable cells was determined using a (CDKs), stage-specific cyclins and inhibi- hemocytometer based on the exclusion tors of cyclin-dependent kinases (CKIs) of trypan blue dye. Twenty four hours act as regulators of the cell cycle.(17) after seeding, the cells were treated with Progression from S to G2-M phase is C. asiatica extract at various concentra- regulated by the accumulation of cyclin tions. B1, which binds to the CDK1 catalytic subunit to form an active complex.(18) Cell proliferation assay In this study, we examined the effects Cell proliferation was measured using of C. asiatica extract on the growth of the MTS [3-(4,5-dimethylthiazol-2-yl)- Caco-2 cells including cell proliferation, 5-(3-carboxymethoxyphenyl)-2-(4- cell cycle, and cyclin B1 protein level to sulfophenyl)-2H-tetrazolium, inner salt] elucidate the mechanism(s) of antitumori- cell proliferation kit (Promega, Corp., genesis of this plant. Madison, WI, USA), which colorimetri- Centella asiatica extract induces cell cycle arrest 23 cally measures as a purple formazan described previously.(19) Briefly, cells were compound produced by viable cells. harvested by centrifugation at 1,000 x g Cells were plated at 104 cells/well in 96- for 5 min at 4 oC. The cell pellets (3x106 well culture plates and treated with 1-16 cells) were washed once with ice-cold mg/mL of C. asiatica extract for 24, 48, PBS and resuspended with 100 µL of the 72 and 96 hr. Then, 20 µL of MTS solu- chilled lysis buffer containing 20 µg/mL tion was added to each well according to leupeptin, 20 µg/mL aprotinin and 0.2 the manufacturer’s instructions. The mix mM phenylmethylsulfonylfluoride tures were incubated for 1 hr at 37 oC and (PMSF). The cells were disrupted by the formazan produced was measured passing them through a G27 needle 10 using a microplate reader spectrophoto- times. After centrifugation in a micro- meter (Corona Electric, Tokyo, Japan) at centrifuge at 750xg for 5 min at 4 oC, the 490 and 630 nm. The results were ex- supernatants were centrifuged again at pressed by the number of cells. 15,000xg for 15 min at 4 oC. The supernatants were divided into aliquots and Flow cytometric analysis of the cell stored at -20 oC. Protein concentration cycle was determined using a Coomassie protein To examine the effect of C. asiatica assay kit (Pierce, Rockford, USA) ac- extract on cell cycle phase distribution, cording to the manufacturer’s instruc- the cells were treated with the extract at tions. Samples were subjected to 12% concentrations of 2-8 mg/mL for various SDS (sodium dodecyl sulfate)-polyacry- times and then resuspended in 500 µL of lamide gel (Wako) electrophoresis with phosphate-buffered saline (PBS) with 200 V for 35 min and transferred to 35% ethanol at 37 oC for 30 min. The polyvinylidene difluoride (PVDF) mem- cells were washed 2 times with PBS and branes (Bio-Rad, Hercules, CA, USA) resuspended in 500 µl of PBS containing with 20 V for 1 h. The membranes were 2 µL of RNaseA (1 mg/mL, Wako, Osaka, blocked in Tween 20-Tris-buffered Japan) and 2 µL of RNaseT1 (3340 saline (TTBS) containing 2% bovine units/mL, Wako), and incubated at 37 oC serum albumin (BSA) for 1 hr and for 1 hr. They were stained with 50 µg/ probed overnight with a primary antibody mL of propidium iodide (PI) and fluores- mouse anti-cyclin B1 (1:2000; Upstate cence was detected with a flow cytome- Biotechnology, Lake Placid, NY) at 4 oC. ter (Coulter Epics XL-MCL, Beckman After washing with TTBS, the mem- Coulter, Tokyo, Japan). The Phoenix branes were incubated further with prima- Flow System was used to make DNA ry antibody mouse anti-β-actin (1:5000; content frequency histograms and Sigma) for 1 hr at 37 oC. Primary antibo- analyze the data. dy binding was detected with a goat anti- mouse IgG conjugated with alkaline phos- Western blot analysis phatase (1:2000; Sigma) and visualized Whole cell lysates were prepared as by an enhanced chemiluminescence

23 24 Bunpo P, et al. method using disodium-3(4-methoxyspi- asiatica extract at concentrations of 2, 4 ro{1,2-dioxetane-3-2’-(5’-chloro)tri- and 8 mg/mL for 48, 72 and 96 h, respec- cyclo[3.3.1.13,7]decan}-4-yl) (CSPD) tively and then subjected to flow cyto- (Boehringer, Manheim Germany). metric analysis after staining DNA with PI. Histograms of the flow cytometric Results data are shown in Fig. 2. In the control C. asiatca extract reduces the proli- culture, most of the cells were distri- feration of Caco-2 cells buted in the G0/G1 phase (Table 1). In C. asiatica extract significantly reduced contrast, 89% of the cells were arrested in the proliferation rate of Caco-2 cells in the S phase after 96 hr of treatment with concentration- and time-dependent man- 8 mg/mL of the extract. The number of ners (Fig. 1). Caco-2 cells treated with Caco-2 cells distributed in the S phase the extract grew very slowly, and the increased in a time-dependent manner at number of cells treated with 1 mg/mL of 2, 4 and 8 mg/mL of the extract, at 48, 72 C. asiatica for 96 hr was approximately and 96 hr. The induction of S phase arrest one-fifth of the medium control. was clearly dose-dependent at 96 h. Sub- G1 peak, which means the appearance of Effect of C. asiatica extract on the apoptotic cells, was observed slightly at cell cycle concentrations of 4 and 8 mg/mL after Caco-2 cells were treated with C. 72 and 96 hr exposure respectively.

* 25 ) 4 20

10 Cell number (1x10 * 5

0 0 20406080100 Time (h)

Figure 1. Effect of C. asiatica extract on the growth of Caco-2. Cells of 1x104 were cultured with or without C. asiatica extract (open circles) at concentrations of 1 (closed circles), 2 (open triangles), 4 (closed triangles), 8 (open squares), 16 (closed squares) mg/mL for 48, 72 and 96 hr. Cell numbers were determined by the MTS assay using various concentrations of the cells as standard. Values are means±SD (n=3). * Significant difference compared with all test concentrations of C. asiatica extract at the same time, p< 0.0001 (ANOVA). Centella asiatica extract induces cell cycle arrest 25

These results indicate that C. asiatica C. asiatica extract induces cyclin B1 extract induced S and G2-M phase accumulation arrests in Caco-2 cells and led to Fig. 3 shows the level of cyclin B1 apoptosis induction in consequence. protein in Caco-2 cells treated with solvent or 2, 4 and 8 mg/mL of the extract for 96 h. Cyclin B1 accumulated distinctively A. 72 h B. 96 h at a concentration of 8 mg/mL in Caco-2 cells. control control Discussion C. asiatica is a plant used for tradi- tional medicine. Crude extract of this

2N 4N 2N 4N plant has been shown to inhibit the proliferation of ascites tumor cells in vitro 2 mg/ml 2 mg/ml and retard the development of transplanted tumors.(9) Several different classes of food constituents and some phytochemicals have been reported to enable modu-

2N 4N 2N 4N lation of both cellular proliferation and programmed cell death.(20) In this study, 4 mg/ml 4 mg/ml we investigated the inhibitory effect of the crude water extract of this plant on cell proliferation and the cell cycle distribution of the colon adenocarcinoma cell line. C. asiatica extract inhibited the 2N 4N 2N 4N proliferation of Caco-2 cells in dose- and 8 mg/ml 8 mg/ml time-dependent manners. Cell cycle analysis showed that Caco-2 cells were preferentially arrested at the S phase. Apoptosis was observed at concentra-

2N 4N 2N 4N tions of 4 and 8 mg/mL after 72 and 96 hr respectively exposures, indicating that the extract reduced the proliferation of Figure 2. Flow cytometric analysis of the cell both cells via cell cycle arrest, and led to cycle. Caco-2 cells were treated with C. asiatica apoptosis. Irreversible arrest of the cell extract at concentrations of 2, 4 and 8 mg/mL for (21,22) 48, 72, 96 h, respectively. The cell number was cycle leads to apoptosis, and it may plotted against DNA content measured in terms be a possible mechanism underlying the of PI fluorescence. The location of diploid (G0- inhibition of cancer cell growth. G1) and tetraploid (G2-M) DNA is indicated by Although the majority of cells in 2N and 4 N. Arrows indicate sub-G1 population. Caco-2 still accumulated in the S phase

25 26 Bunpo P, et al.

Table 1. Cell cycle analysis of Caco-2 cells. Percentage distribution of cells treated with C. asiatica extract at doses of 2, 4 and 8 mg/mL. Concentration Percentage of cells at (mg/mL) G0-G1 S G2-M 72 hr 96 hr 72 hr 96 hr 72 hr 96 hr 0 607526151410 2 504829332119 4 472525472828 8 29061891011

and a number of other regulatory pro- Cyclin B1 (58 kDa) teins are degraded by APC-mediated specific proteolysis during and after β-actin mitosis. Proteolysis of cyclin B1 is uni- 0 2 4 8 versally required as an exit from mitosis. (16) C.asiatica mg/mL In our study, cyclin B1 protein levels in Caco-2 cells were distinctly higher Figure 3. Effect of C. asiatica extract on the levels after treatment with the extract. There- of cyclin B1 in Caco-2 cells. The cells were fore, C. asiatica extract affects the cell treated with and without C. asiatica extract at 2, cycle through accumulation of cyclin 4 and 8 mg/mL for 96 hr. Forty µg protein from B1, probably by modification of APC each sample were electrophoresed and Western blot was carried out using antibody against cyclin activity or its regulation. Although lower B1, and β-actin was detected as a loading control. concentrations of C. asiatica extract, at less than 1 mg/mL, did not exhibit any at 96 hr, some were arrested at the G2-M significant cytotoxicity on Caco-2 cells phase. This difference in response may (data not shown), this might be improved be due to the possible effect of several with a long period of incubation. The differently-active components in the water extract used in this study probably crude extract of C. asiatica. Cell cycle is contained various kinds of constituents, regulated by oscillating changes in the (11) partially purification of the extract, activity of the cell cycle machinery, and which might may reduce the concentra- specific proteolysis of the cell cycle tion used in the cell culture. regulators.(16) Progression from the S to These results indicated that C. asia- G2-M phase is regulated by the accumu- tica extract inhibits the proliferation of lation of cyclin B1, which binds the to colon cancer cells through modification cdk1 catalytic subunit to form an active of cell cycle progression. C. asiatica is complex.(18) During mitosis, a specific usually taken as a tnaditional medicine ubiquitin-ligase complex, known as an in Thailand. Therefore, the detailed anaphase-promoting complex (APC), is mechanisms of C.asiatica cell cycle activated to initiate proteolysis of various arrest and it other possible modes of mitosis regulators. Cyclin B1, cyclin A action should be clarified before intro- Centella asiatica extract induces cell cycle arrest 27 ducing this plant or its active compo- formation in male F344 rats. J Med Invest nents to clinical use. 2002;49:25-34. 7. Shukla A, Rasik AM, Dhawan BN. Asiatico- side-induced elevation of antioxidant levels Acknowledgements in healing wounds. Phytother Res 1999;13: P. Bunpo is a graduate student supported by 50-4. the AIEJ (Association of International Education, 8. Yoosook C, Bunyapraphatsara N, Boonya- Japan) Short-term Student Exchange Promotion kiat Y, Kantasuk C. Anti-herpes simplex Program Scholarship and the 2000 Royal Golden virus activities of crude water extracts of Jubilee Ph.D. Research Assistantship. Thai medicinal plants. Phytomedicine 2000; 6:411-9. References 9. Babu TD, Kuttan G, Padikkala J. Cytotoxic 1. Lee K-H. Current development in the dis- and anti-tumour properties of certain taxa of covery and design of new drug candidates Umbelliferae with special reference to Cen- from plant natural product leads. J Nat Prod tella asiatica (L.) Urban J Ethnopharmacol 2004;67:273-83. 1995;48:53-7. 2. da Rocha AB, Lopes RM, Schwartsmann G. 10. Bunpo P, Kataoka K, Arimochi H, et al. Natural products in anticancer therapy. Curr Inhibitory effects of Centella asiatica on Opin Pharmacol 2001;1:364-9. azoxymethane-induced aberrant crypt focus 3. Suaeyun R, Kinouchi T, Arimochi H, formation and carcinogenesis in the intes- Vinitketkumnuen U, Ohnishi Y. Inhibitory tines of F344 rats. Food Chem Toxicol 2004; effects of lemon grass (Cymbopogon citratus 42:1987-1997. Stapf) on formation of azoxymethane- 11. Inamdar PK, Yeole RD, Ghogare AB, Souza induced DNA adducts and aberrant crypt NJ. Determination of biologically active foci in the rat colon. Carcinogenesis 1997; constituents in Centella asiatica. J Chroma- 18:949-55. togr A 1996;742:127-30. 4. Chewonarin T, Kinouchi T, Kataoka K, et 12. Renehan AG, Bach SP, Potten CS. The al. Effects of roselle (Hibiscus sabdariffa relevance of apoptosis for cellular homeo- Linn.), a Thai medicinal plant, on the muta- stasis and tumorigenesis in the intestine. Can genicity of various known mutagens in J Gastroenterol 2001;15:166-76. Salmonella typhimurium and on the forma- 13. Mori H, Sugie S, Yoshimi N, Hara A, tion of aberrant crypt foci induced by the Tanaka T. Control of cell proliferation in colon carcinogens azoxymethane and 2- cancer prevention. Mutat Res 1999;428: amino-1-methyl-6-phenylimidazole (4,5-b) 291-8. pyridine in F344 rats. Food Chem Toxicol 14. Singh RP, Dhanalakshmi S, Agarwal R. 1999;37:591-601. Phytochemicals as cell cycle modulators--a 5. Chiampanichayakul S, Kataoka K, Arimochi less toxic approach in halting human cancers. H, et al. Inhibitory effects of bitter melon Cell Cycle 2002;1:156-61. (Momordica charantia Linn.) on bacterial 15. Johnson IT. Anticarcinogenic effects of diet- mutagenesis and aberrant crypt focus forma- related apoptosis in the colorectal mucosa. tion in the rat colon. J Med Invest 2001;48: Food Chem Toxicol 2002;40:1171-8. 88-96. 16. King RW, Deshaies RJ, Peters J-M, 6. Intiyot Y, Kinouchi T, Kataoka K, et al. Kirschner MW. How proteolysis drives the Antimutagenicity of Murdannia loriformis cell cycle. Science 1996;274:1652-9. in the Salmonella mutation assay and its 17. Frouin I, Montecucco A, Biamonti G, inhibitory effects on azoxymethane-induced Hubscher U, Spadari S, Maga G. Cell cycle- DNA methylation and aberrant crypt focus dependent dynamic association of cyclin/Cdk

27 28 Bunpo P, et al.

complexes with human DNA replication 20. Johnson IT. Anticarcinogenic effects of diet- proteins. EMBO J 2002;21:2485-95. related apoptosis in the colorectal mucosa. 18. Doree M, Hunt T. From Cdc2 to Cdk1: Food Chem Toxicol 2002;40:1171-8. when did the cell cycle kinase join its cyclin 21. Makin G, Hickman JA. Apoptosis and partner? J Cell Sci 2002;115:2461-4. cancer chemotherapy. Cell Tissue Res 2000; 19. Yang J, Liu X, Bhalla K, et al. Prevention of 301:143-52. apoptosis by Bcl-2: release of cytochrome c 22. Singh RP, Dhanalakshmi S, Agarwal R. from blocked mitochondria. Science 1997; Phytochemicals as cell cycle modulators. A 275:1129-32. less toxic approach in halting human cancers. Cell cycle 2002;1:156-61.

สารสกัดบัวบกเหนี่ยวนํ าการหยุดวัฏจักรของเซลลมะเร็งลํ าไสใหญของคนชนิด Caco-2

ปยะวรรณ บุญโพธิ์,1 วท.บ., Keiko Kataoka,2 Ph.D., Hideki Arimochi,2 Ph.D., Haruyuki Nakayama,2 M.Sc., Tomomi Kuwahara,2 Ph.D., Yoshinari Ohnishi,2 Ph.D., อุษณีย วนิ ิจเขตคํ านวณ,1 Ph.D. 1ภาควชาชิ ีวเคมี คณะแพทยศาสตร มหาวิทยาลัยเชียงใหม ประเทศไทย, 2Department of Molecular Bacteriology, Graduate School of Medicine, The University of Tokushima, Tokushima, Japan 770-8503

บทคัดยอ บัวบกจัดเปนพืชสมุนไพรซึ่งนํ ามาใชกันมากในทางการแพทยแผนโบราณของไทย การ ศึกษาครั้งนี้เพื่อตรวจหาฤทธิ์ตานมะเร็งของสารสกัดนํ้ าบัวบกโดยใชเซลลมะเร็งลํ าไสใหญซึ่งได มาจากมนุษยชนิด Caco-2 ผลการศึกษาพบวาสารสกัดบัวบกลดการแบงตัวของเซลลมะเร็งอยาง แปรผันตรงตามความเขมขนและเวลาอยางมีนัยสํ าคัญ กลไกการยับยั้งการเจริญของเซลลมะเร็ง เกิดขึ้นผานทางการหยุดวัฏจักรเซลล ซึ่งสารสกัดบัวบกเหนี่ยวนํ าใหเซลลมะเร็งชนิด Caco-2 หยุด

การเจริญไวที่ระยะ S และ G2-M และพบวามีการตายของเซลลเกิดขึ้นรวมกัน สารสกัดบัวบกเพิ่ม การสะสมของโปรตีน cyclin B1 ภายในเซลล การศึกษานี้บงชี้วาสารสกัดบัวบกลดการแบงตัว ของเซลลมะเร็งลํ าไสใหญชนิด Caco-2 ผานทางกระบวนการในวัฏจักรเซลลและการหยุดวัฏจักร เซลลนี้มีความเกี่ยวของกับการสะสมของโปรตีน cyclin B1 เชียงใหมเวชสาร 2548;44(1):21-28. คํ าสํ าคัญ: บัวบก เซลลมะเร็งลํ าไสใหญ การแบงตัวของเซลล การหยุดวัฏจักรเซลล cyclin B1