Neuroprotective Effects of Constituents of the Oriental Crude Drugs, Rhodiola Sacra, R

August 2002 Notes Biol. Pharm. Bull. 25(8) 1101—1104 (2002) 1101

Neuroprotective Effects of Constituents of the Oriental Crude Drugs, Rhodiola sacra, R. sachalinensis and Tokaku-joki-to, against Beta-amyloid Toxicity, Oxidative Stress and Apoptosis

a a c c c Inhee MOOK-JUNG, Hee KIM, Wenzhe FAN, Yasuhiro TEZUKA, Shigetoshi KADOTA, d ,b Hisao NISHIJO, and Min Whan JUNG* a Brain Disease Research Center, Ajou University School of Medicine; b Institute for Medical Sciences, Ajou University School of Medicine; Suwon 442–721, Korea: c Institute of Natural Medicine, Toyama Medical and Pharmaceutical University; and d Faculty of Medicine, Toyama Medical and Pharmaceutical University; 2630 Sugitani, Toyama 930–0194, Japan. Received February 1, 2002; accepted April 12, 2002

We tested the constituents of two Rhodiola plants, Rhodiola sacra S. H. FU and R. sachalinensis A. BOR, and an Oriental crude drug, Tokaku-joki-to, for their neuroprotective effects. Of the 58 compounds tested, six had considerable protective effects against beta-amyloid-induced death of B103 neuronal cells in vitro. These six compounds also showed protective effects against staurosporine-induced cell death, and two of the six compounds

protected neurons from H2O2-induced cell death. These results suggest that some of the tested compounds protect neurons from beta-amyloid toxicity based on antiapoptotic and antioxidative activity. Key words beta-amyloid; apoptosis; oxidative stress; natural medicine; Alzheimer’s disease; B103 cell

Alzheimer’s disease (AD) is a neurodegenerative disorder from two Rhodiola plants, R. sacra and R. sachalinensis, and currently without an effective treatment. Impairment of a Kampo formula Tokaku-joki-to (Kotaro Kampo, Japan), as memory is the initial and most signiﬁcant symptom of AD. previously reported.10—12) Hydroquinone (1), 4-hydroxyben- Although the exact cause of the disease is yet to be deter- zoic acid (2), b-D-glucopyranosyl 4-hydroxybenzoic acid (3), mined, the weight of current evidence suggests strongly that protocatechuic acid (4), gallic acid (5), 4-hydroxycinnamic the neurotoxic action of overproduced beta-amyloid (Ab) is acid (8), caffeic acid (9), 2-phenylethyl b-D-glucopyranoside 1,2) the prime cause of AD. According to this hypothesis, ﬁnd- (13), 2-phenylethyl a-L-arabinopyranosyl-(1→6)-b-D-glu- ing compounds that can block the neurotoxic action of Ab is copyranoside (14), rhodiocyanoside A (17), sarmentosin an important avenue toward developing new treatments for (18), heterodendorin (20), sacranoside A (29), 3-O-gal- AD. One approach is examining natural compounds that have loylepigallocatechin-(4b→8)-epigallocatechin 3-O-gallate been extracted from traditional drugs that are known to have (35) and (Ϫ)-epigallocatechin-3-O-gallate (39) were isolated nootropic functions. The memory enhancing effects of some from R. sacra, and compounds 5, salidroside (6), 8, p-tyrosol of these drugs are possibly attributable to their actions (10), 6Љ-O-galloylsalidroside (11), benzyl b-D-glucopyra- against Ab neurotoxicity. noside (12), 13, trans-cinnamyl b-D-glucopyranoside (15), In the present study we examined effects of natural com- rosarin (16), 17, lotaustralin (19), octyl b-D-glucopyranoside pounds that were isolated from two Rhodiola plants, Rhodi- (21), 1,2,3,6-tetra-O-galloylglucose (22), 1,2,3,4,6-penta-O- ola sacra S. H. FU and R. sachalinensis A. BOR (Crassu- galloylglucose (23), kaempferol (24), kaempferol 3-O-b-D- laceae), and a Kampo formula Tokaku-joki-to (Persica and xylopyranosyl(1→2)-b-D-glucopyranoside (25), kaempferol Rhubarb Combination). Tokaku-joki-to is an Oriental blood- 3-O-b-D-glucopyranosyl(1→2)-b-D-glucopyranoside (26), quickening and stasis-transforming prescription, which con- rhodionin (27), rhodiosin (28), rosiridin (30), sachalinoside sists of Persicae Semen, Rhei Rhizoma, Cinnamomi Cortex, A (31), sachalinol A (32), sachalinol C (33), sachalinol B Glycyrrhizae Radix and Mirabilitum. It is clinically used in (34), 35 and (Ϫ)-epigallocatechin (37) were isolated from R. the People’s Republic of China and Japan for treating diverse sachalinensis. Compounds 4, 5, cinnamic acid (7) and (Ϫ)- symptoms, including amenorrhea, climacteric syndrome and epicatechin 3-O-gallate (40), licuroside (41), liquiritigenin 3,4) emotional imbalance. Rhodiola plants grow in the Chang- (42), liquiritigenin 4Ј-O-b-D-glucopyranoside (43), (Ϫ)-epi- bai Mountain area, Tibet and Xinjiang autonomous regions catechin (44), torachrysone 8-O-b-D-glucopyranoside (45), in China. A Tibetan folk medicine, Rhodiola Radix, is pre- emodin (46), emodin 8-O-b-D-glucopyranoside (47), trans- pared from several alpine Rhodiola plants and is used as a 3,5,4Ј-trihydroxystilbene 4Ј-O-b-D-glucopyranoside (48), 5) hemostatic, tonic and contusion. Recently, it has been re- trans-3,5,4Ј-trihydroxystilbene 4Ј-O-b-D-(2-O-galloyl)glu- 6) ported that Rhodiola plants improve learning and memory. copyranoside (49), trans-3,5,4Ј-trihydroxystilbene 4Ј-O-b-D- In spite of extensive studies on their activities,5—9) their ef- (6-O-galloyl)glucopyranoside (50), cis-3,5,4Ј-trihydroxystil- fects on neuronal cell death have not been tested. Here we bene 4Ј-O-b-D-(6-O-galloyl)glucopyranoside (51), gallic acid examined the effects of compounds that were extracted from 4-O-b-D-(6-O-galloyl)glucopyranoside (52), 1-O-galloylglu- two Rhodiola plants and Tokaku-joki-to against Ab-induced cose (53), 4-(4-hydroxyphenyl)-2-butanone 4Ј-O-b-D-glu- neuronal cell death, oxidative stress and apoptosis. copyranoside (54), isolindleyin (55), lindleyin (56), 1,2,6-tri- O-galloylglucose (57), 4-(4-hydroxyphenyl)-2-butanone 4Ј- MATERIALS AND METHODS O-b-D-(2-O-galloyl-6-O-cinnamoyl)glucopyranoside (58) were isolated from Tokaku-joki-to (Fig. 1). Note that some Fifty-six compounds (1—35, 37, 39—58) were isolated compounds were isolated from more than one source. Two

Fig. 1. Structures of 58 Compounds (1—58) Used in This Study August 2002 1103 compounds, (ϩ)-catechin (36) and (ϩ)-gallocatechin (38), Table 1. Effects of Isolated Compounds on Ab Toxicity were purchased from Funakoshi (Tokyo, Japan). Rat brain-derived B103 cells, which do not express en- Cell survival Cell survival 13) dogenous amyloid precursor protein, were used in the pre- Control 100Ϯ2.5 29 50.9Ϯ2.2 Ϯ Ϯ sent study. B103 cells were grown in Dulbecco’s modified Ab 25—35 63.5 2.6 30 55.3 4.2 Eagle’s medium (DMEM) containing 10% fetal bovine 1 59.7Ϯ8.0 31 59.4Ϯ4.6 2 50.9Ϯ3.7 32 53.2Ϯ2.2 serum and 5% penicillin/streptomycin. At the outset, 90% Ϯ Ϯ ϫ 3 3 49.9 4.3 33 59.4 6.6 confluent cells were dissociated and plated at 5 10 4 52.1Ϯ5.2 34 53.6Ϯ3.6 cells/well in a 96-well plate. When the cells were attached to 5 59.7Ϯ5.5 35 60.2Ϯ5.1 the plate, the medium was replaced with plain DMEM. Cell 6 53.5Ϯ6.5 36 66.8Ϯ3.1 death was induced by adding Ab , H O , or stau- 7 50.6Ϯ1.1 37 62.1Ϯ6.1 25—35 2 2 Ϯ Ϯ rosporine in the culture for 24 h. Ab was prepared as a 8 54.6 2.8 38 66.1 1.2 25—35 9 56.7Ϯ2.4 39 71.0Ϯ0.7 10 mM stock solution in dimethyl sulfoxide (DMSO) and di- 10 57.6Ϯ1.8 40 86.0Ϯ2.5 luted with phosphate buffered saline to induce aggregation. 11 60.5Ϯ1.6 41 63.4Ϯ0.5 Ϯ Ϯ H2O2 was prepared as a 30% stock solution in water and 12 51.8 1.4 42 60.8 0 freshly diluted with water before each use. Staurosporine was 13 57.3Ϯ3.4 43 63.5Ϯ4.4 14 54.9Ϯ0.1 44 77.5Ϯ0.9 prepared as a 1 mM stock solution in DMSO and diluted with 15 59.6Ϯ1.3 45 66.7Ϯ2.5 water to make a 500 nM working solution. The extracted 16 52.4Ϯ0.8 46 67.6Ϯ0.7 compounds were treated 1 h before adding each toxicant to 17 59.9Ϯ1.0 47 57.3Ϯ1.6 the medium. The final concentration of each compound was 18 55.4Ϯ1.1 48 64.9Ϯ0.5 Ϯ Ϯ 10 m M throughout the experiments. Ab was purchased 19 47.9 0.2 49 79.6 8.0 25—35 20 55.9Ϯ1.6 50 77.2Ϯ10.6 from US Peptide (Fullerton, CA, U.S.A.). DMEM, fetal 21 55.4Ϯ0.1 51 77.7Ϯ6.5 bovine serum and penicillin/streptomycin were purchased 22 77.7Ϯ7.6 52 59.4Ϯ6.4 from GibcoBRL (Grand Island, NY, U.S.A.). All other chem- 23 68.9Ϯ4.9 53 72.6Ϯ7.5 icals were purchased from Sigma (St. Louis, MO, U.S.A.). 24 61.8Ϯ1.1 54 58.1Ϯ4.7 B103 cells were generously provided by Dr. David Schubert 25 64.5Ϯ1.7 55 72.9Ϯ6.6 26 47.9Ϯ2.7 56 65.2Ϯ4.8 (Salk Institute, La Jolla, CA, U.S.A.). 27 61.7Ϯ1.9 57 60.1Ϯ4.2 The degree of cell survival was quantified using 3-(4,5-di- 28 61.5Ϯ1.6 58 70.5Ϯ2.9 methylthiazol-2-yl)-2,5-dimethyltetrazolium bromide (MTT) 14) B103 cells were treated with 25 m M Ab for 18 h, together with one of 58 differ- assay. At the end of cell culture, MTT was added to the 25—35 ent compounds (1—58), and the degree of cell survival was assessed using MTT assay. culture (1 : 10 v : v; MTT solution/culture medium) and incu- The values indicate the degree of cell survival expressed as a % of the control level bated for 4 h at 37 °C. Cells were then solubilized in 50% di- (meanϮS.E.M.). methylformamide and 10% sodium dodecyl sulfide (pH 4.7). The degree of cell survival was determined based on the ab- tested, only two enhanced cell survival compared to the sorbance measured at O.D. 570—630 nm using a plate reader H2O2-treated group. Compounds 22 and 44 enhanced cell (Bio-Tek Instruments, Winooski, VT, U.S.A.). The effect of survival to 81.6Ϯ1.7% and 76.7Ϯ0.9% of the control level, each extract on cell death was assessed in quadruplicate, and respectively. These amount to 39.9% and 23.9% blockade of Ϯ the data are expressed as the mean S.E.M. H2O2-induced cell death, respectively. The effects of the six compounds on H2O2-induced cell death are shown in Fig. RESULTS AND DISCUSSION 2A. Staurosporine is known to induce apoptotic cell death.13)

Following treatment of 25 m M Ab 25—35 for 18 h, cell sur- Treatment with 50 nM staurosporine for 18 h reduced cell survival was reduced to 63.5Ϯ2.6% of the control culture vival to 37.2Ϯ0.5% of the control culture (62.8% cell death). (36.5% cell death). Of the 58 compounds that were co- All six compounds that showed protective effects against Ab treated with Ab, six (22, 40, 44, 49—51) showed more than toxicity enhanced cell survival when treated together with 10% enhancement of cell survival at 10 m M compared to the staurosporine (Fig. 2B). The degrees of cell survival were be- Ab treatment group. Ten % enhancement of cell survival tween 50.3Ϯ0.8% and 63.6Ϯ1.1% of the control level compared to control level amounts to ca. 27% blockade of (20.9—42.0% blockade of staurosporine-induced cell death). Ab-induced cell death. The degrees of cell survival ranged The main objective of the present study was to test from 77.2Ϯ6.5 to 86.0Ϯ2.5% of the control level (37.5— whether compounds isolated from R. sacra, R. sachalinensis 61.6% blockade of Ab-induced cell death). This is compara- and Tokaku-joki-to protect neurons against beta-amyloid tox- ble to the level of cell protection by estrogen (ca. 36% block- icity. Of the 58 tested compounds, six (22, 40, 44, 49—51) ade of Ab-induced cell death) under a similar experimental protected B103 cells from Ab-induced cell death in a consid- condition.20) Table 1 shows the effects of all 58 compounds erable manner (enhancement Ͼ10% of the control). The na- on Ab-induced B103 cell death. ture of Ab-induced cell death is not entirely clear. Previous We further tested the effects of the six compounds that studies suggest that apoptotic cell death, oxidative stress and showed substantial protective effects against Ab toxicity, necrotic cell death are all involved in Ab-induced cell against oxidative stress and apoptosis. Oxidative stress was death.15—20) In the present study, we examined the antioxida- induced by treating the cells with 200 m M of H2O2 for 18 h. tive and antiapoptotic effects of the six compounds that pro- Ϯ H2O2 treatment reduced cell survival to 69.4 1.6% of the tected neurons from Ab toxicity. The results show that all of control culture (30.6% cell death). Of the six compounds them had antiapoptotic effects, whereas two compounds (22, 1104 Vol. 25, No. 8

Acknowledgments This work was supported by the Korea Research Foundation Grant 1998-019-F00059 to M.W. J.

REFERENCES

1) Cordell B., Annu. Rev. Pharmacol. Toxicol., 34, 69—89 (1994). 2) Selkoe D. J., J. Neuropathol. Exp. Neurol., 53, 438—447 (1994). 3) Hsu H., William G., “Shang han lun (The Great Classic of Chinese Medicine),” Oriental Healing Arts Institute, Los Angeles, 1981, p. 18. 4) Terasawa K., “Kampo Japanese-Oriental Medicine Insight from a Clinical Case,” K. K. Standard McIntyre, Tokyo, 1993, p. 236. 5) Yang Y. C., He T. N., Lu S. L., Hung R. F., Wang Z. X., “Zangyao Zhi (Manual of Tibetan Materia Medica),” Qinghai People’s Publishing House, Xining, 1991, pp. 432—434. 6) Ming H. Q., Xia G. C., Zhang R. D., Zhongcaoyao, 19, 229—234 (1988). 7) Yang Z. Y., Liu Q., Zhong C. S., Sun S. L., Jiang B. R., Xiao Y. W., Zhongcaoyao, 26, 441—442 (1995). 8) Li J. X., Liu J. T., Jin Y. R., Zhang H. G., Wu G. X., Okuyama T., Zhongcaoyao, 29, 659—661 (1998). 9) Liu C. B., Jin Y., Li N., Xie J., Xu J. F., Tianran Chanwu Yianjiu Yukaifa, 11, 18—22 (1999). 10) Fan W., Tezuka Y., Komatsu K., Namba T., Kadota S., Biol. Pharm. Fig. 2. Effects of Isolated Compounds on Oxidative Stress-Induced and Bull., 22, 157—161 (1999). Apoptotic Cell Death 11) Fan W., Tezuka Y., Ni D. K. M., Kadota S., Chem. Pharm. Bull., 49, B103 cells were treated with 200 m M of H2O2 (A) or 50 nM staurosporine (B) for 18 h, 396—401 (2001). together with one of the six compounds that showed substantial protective effects against Ab toxicity, and the degree of cell survival was assessed using MTT assay. The 12) Fan W., Tezuka Y., Kadota S., Chem. Pharm. Bull., 48, 1055—1061 values indicate the degree of cell survival expressed as a % of the control level. Error (2000). bars indicate S.E.M. STS: staurosporine. 13) Schubert D., Heinemann S., Carlisle W., Tarikas H., Kimes B., Patrick J., Steinbach J. H., Culp W., Brandt B. L., Nature (London), 249, 224—227 (1974). 44) had additional antioxidative effects. Thus, the neuropro- 14) Behl C., Davis J. B., Klier F. G., Schubert D., Brain Res., 645, 253— tective effects of the six compounds against Ab toxicity are 264 (1994). probably based on both antiapoptotic and antioxidative ef- 15) Loo D. T., Agata C., Pike C. J., Whittemore E. R., Wulencewicz A. J., fects, but antiapoptotic effects appear to play a more impor- Cotman C. W., Proc. Natl. Acad. Sci. U.S.A., 90, 7951—7955 (1993). tant role. 16) Behl C., Davis J. B., Lesley R., Schubert D., Cell, 77, 817—827 (1994). It was difﬁcult to assess the structure–activity relationship 17) Schubert D., Behl C., Lesley R, Brack A., Dargusch R., Sagara Y., of the tested compounds based on the current results. Future Kmura H., Proc. Natl. Acad. Sci. U.S.A., 92, 1989—1993 (1995). studies are required for clear elucidation of structures that are 18) Green P. S., Gridley K. E., Simpkins J. W., Neurosci. Lett., 218, 165— responsible for the neuroprotective effects. Improvement of 168 (1996). 19) Mook-Jung I., Joo I., Sohn S., Kwon H. J., Huh K., Jung M. W., Neu- neuroprotective effects by synthetic modiﬁcation based on rosci. Lett., 235, 101—104 (1997). the structure–activity relationship may lead to development 20) Kim H., Bang O. Y., Jung M. W., Ha S. D., Hong H. S., Huh K., Kim of new drugs for AD. S. U., Mook-Jung I., Neurosci. Lett., 302, 58—62 (2001).