- Email: [email protected]
Novel antioxidants isolated from plants of the genera Ferula, Inula, Prangos and Rheum collected in Uzbekistan
ARTICLE IN PRESS
Phytomedicine 11 (2004) 645–651 www.elsevier.de/phymed
Novel antioxidants isolated from plants of the genera Ferula, Inula, Prangos and Rheum collected in Uzbekistan K. Kogure, I. Yamauchi, A. Tokumura, K. Kondou, N. Tanaka, Y. Takaishi, K. Fukuzawa Faculty of Pharmaceutical Sciences, University of Tokushima, Shomachi-1, 770-8505 Tokushima, Japan Received 7 March 2003; accepted 5 September 2003
Abstract We examined the effects of 48 compounds isolated from Ferula pallida, F. penninervis, Inula macrophylla, Prangos pabularia, P. tschimganica and Rheum maximowiczii collected in Uzbekistan on ADP/Fe2+-induced lipid peroxidation of egg yolk phosphatidylcholine liposomes. Of those compounds, 23 inhibited ADP/Fe2+-induced lipid peroxidation and nine showed especially strong inhibition of lipid peroxidation. Most compounds that inhibited peroxidation scavenged the 1,10 -diphenyl-2-picrylhydrazyl (DPPH) radical, indicating that the inhibition was due to radical scavenging. However, some compounds did not scavenge DPPH but inhibited lipid peroxidation significantly, suggesting that their inhibitory effect was not due to radical scavenging but to some other mechanism, such as prevention of Fe2+ function. Thus, we found various new antioxidants, some of which had a unique mechanism of action, in Ferula, Inula, Prangos and Rheum plants collected in Uzbekistan as seeds used in medicine. r 2004 Elsevier GmbH. All rights reserved. Keywords: Antioxidants; Natural products; Ferula; Inula; Prangos; Rheum
Introduction Oxidative stresses are suggested to be related to various diseases and pathological conditions such as atherosclerosis (Aviram, 2000), carcinogenesis (Hsu et al., 2000) and aging (Cadenas and Davies, 2000). Thus, to eliminate oxidative stresses such as reactive oxygen species in the body, many workers have been attemptings to develop antioxidative drugs (Simonini et al., 2000). We have studied the antioxidation effects of various compounds, such as the triterpene celastrol (Sassa et al., 1990, 1994), the bisbenzylisoquinoline alkaloid cepharanthine (Kogure et al., 1999), the Corresponding author. Tel.: +81-88-633-7248; fax:+81-88-6339572. E-mail address: [email protected] (K. Fukuzawa).
0944-7113/$ - see front matter r 2004 Elsevier GmbH. All rights reserved. doi:10.1016/j.phymed.2003.09.004
xanthophyll astaxanthine (Goto et al., 2001), the pungent ingredient of red pepper known as capsaicin (Kogure et al., 2002) and artificial pigment cyanine dyes (Kogure et al., 1998), in order to develop antioxidants for medical use. Recently, we examined traditional crude drugs of central Asian people, and found various seed compounds (Su et al., 2000a, b, 2001; Zhou et al., 2000; Shikishima et al., 2001a-c; Tamemoto et al., 2001; Tada et al., 2002; Shikishima et al., in press). As traditional crude drugs in Uzbekistan have never been examined scientifically, the compounds in these crude drugs are very likely to contain new medicinal compounds. Recently, we reported that compounds isolated from various plants collected in Uzbekistan showed anti-HIV activity in vitro (Shikishima et al., 2001c) and preventive effects on the generation and release of inflammation
ARTICLE IN PRESS 646
K. Kogure et al. / Phytomedicine 11 (2004) 645–651
agents such as TNF-a and IL-2 in vitro (Tada et al., 2002; Shikishima et al., 2002). These biological effects might thus be associated with antioxidative activity. In this study, to find candidates among medicinal seeds from traditional Uzbek crude drugs, we examined the effects of 48 compounds isolated from Ferula pallida, F. penninervis, Inula macrophylla, Prangos pabularia, P. tschimganica and Rheum maximowiczii collected in Uzbekistan on ADP/Fe2+-induced lipid peroxidation of egg yolk phosphatidylcholine liposomes (Fukuzawa et al., 1995).
Assay of radical scavenging The radical scavenging activities of test compounds (20 mM) in the ethanolic solution were determined from the decrease in the absorbance of DPPH radical at 517 nm due to their scavenging of an unpaired electron of the stable DPPH radical in a mixture of 10 ml of ethanol, 10 ml of 50 mM 2-(N-morpholino)ethanesulfonic acid (MES) buffer (pH 5.5) and 5 ml of 0.5 mM DPPH in ethanol (Sassa et al., 1990, 1994; Kogure et al., 1999).
Measurements of absorption spectra
Materials and methods Materials The plants F. pallida, F. penninervis, I. macrophylla, P. pabularia, P. tschimganica and R. maximowiczii were collected in Uzbekistan, and many natural compounds were isolated from them as described previously (Su et al., 2000a b, 2001; Zhou et al., 2000; Shikishima et al., 2001a–c; Tamemoto et al., 2001; Tada et al., 2002; Shikishima et al., in press). Egg yolk phosphatidylcholine (EyPC) was purchased from Nihon Seika Co. (Tokyo). 1,10 -Diphenyl-2-picrylhydrazyl (DPPH) was obtained from Nacalai Tesque (Kyoto). a-Tocopherol was kindly provided by Eisai Co. (Tokyo, Japan). Other reagents were of the highest grade commercially available. The solution of Fe2+ was prepared by dissolving FeSO4 in water just before use, to avoid precipitation of Fe3+ as Fe(OH)3.
Assay of lipid peroxidation Liposomes consisting of EyPC were prepared by sonication in a bath-type sonicator in 10 mM Tris-HCl buffer (pH 7.4). Solutions of the compounds with dimethyl sulfoxide were added to liposomes (0.2 mmol lipid/ml) suspended with 10 mM Tris-HCl buffer (pH 7.4) at 25 1C. Immediately after their addition, 1 mM ADP and 0.1 mM FeSO4 were added to the suspension to induce lipid peroxidation. After 15 min, an ethanolic solution of the well-known antioxidant butylated hydroxytoluene (BHT) was added to the reaction mixture (final concentration, 4.5 mM) to terminate lipid peroxidation. Amounts of lipid peroxides were determined as amounts of thiobarbituric acid-reactive substances (TBARS) in terms of malonedialdehyde (MDA), using tetraethoxypropane as a standard (Goto et al., 2001).
The absorption spectra of test samples in 10 mM TrisHCl buffer (pH 7.4) were measured on a spectrophotometer (Shimadzu UV-1600) at 25 1C.
Results We examined the effects of 48 compounds (chemical structures shown in Fig. 1) isolated from F. pallida, F. penninervis, I. macrophylla, P. pabularia, P. tschimganica and R. maximowiczii collected in Uzbekistan on ADP/Fe2+-induced lipid peroxidation of EyPC liposomes, by comparison with those of the well-known antioxidants cysteine and a-tocopherol. Cysteine and atocopherol inhibited liposomal lipid peroxidation (Table 1). The inhibitory effect of cysteine was potent in this system. As listed in Table 1, 23 compounds (final concentrations, 100 mM) inhibited lipid peroxidation. The inhibition percentages of nine compounds (KT98, 99, 100, 101, 108, 118, 119, 129 and 133) were over 70% as compared to that of cysteine. KT100 and KT118 were required at especially low concentrations for 50% inhibition (IC50 values) (2.5 mM and 0.7 mM, respectively). Among the compounds isolated from F. pallida and P. pabularia, no potent antioxidant was observed. The number of potent antioxidative compounds in R. maximowiczii was the highest. To obtain information about the mechanisms of the antioxidative effects of the compounds, we examined their radical scavenging effects by measuring absorbance change of DPPH radical at 517 nm (Table 1). We used cysteine as a standard, one molecule of which scavenges one DPPH radical (Sassa et al., 1990). In our system, addition of cysteine (final concentration, 20 mM) immediately decreased absorbance at 517 nm by scavenging DPPH radical. Over 2/3 of the compounds that inhibited lipid peroxidation scavenged DPPH radical. The compounds that scavenged most strongly were KT08, 110, 96, 97, 98, 99 and 100. However, some compounds (KT101, 108, 118, 119 and 133) did not
ARTICLE IN PRESS K. Kogure et al. / Phytomedicine 11 (2004) 645–651
647
Fig. 1. Chemical structures of cysteine, a-tocopherol and compounds isolated from Ferula, Inula, Prangos and Rheum plants.
ARTICLE IN PRESS 648
K. Kogure et al. / Phytomedicine 11 (2004) 645–651
Table 1. Values of inhibition percentage of lipid peroxidation and absorbance change due to scavenging of DPPH radical by compounds isolated from Ferula, Inula, Prangos and Rheum plants, and categories of them classified in Fig. 2 Plant
Compound
Inhibition of lipid peroxidation(%)a
Radical scavenging (DA517 )b
Category
F. pallida
KT KT KT KT KT KT KT KT
23 29 31 34 35 36 37 38
16.4 7.7 0 0 0 0 0 0
0.158 0 0 — — — — —
E G G
F. penninervis
KT KT KT KT KT KT KT KT KT KT KT KT KT
103 104 106 107 108 110 111 116 117 118 119 120 121
0 0 0 0 83.9 (IC50 60 mM) 15.7 0 0 0 103.0 (IC50 0.7 mM) 72.1 (IC50 70 mM) 0 0
— — — — 0.004 0.634 — — — 0.130 0.143 — —
C E
C C
I. macrophylla
KT 08 KT 10 KT 11
35.9 5.8 0
0.648 0.232 —
P. pabularia
KT KT KT KT KT KT KT
64 66 68 71 72 74 75
0 0 0 0 4.3 11.8 40.6
— — — — 0.004 0.043 0.069
KT KT KT KT KT KT KT KT KT
123 125 127 129 130 131 132 133 135
28.8 0 2.1 91.7 (IC50 50 mM) 25.7 18.6 0 87.2 (IC50 75 mM) 8.0
0 — 0 0.331 0.044 0 — 0.019 0.082
KT KT KT KT KT KT
96 97 98 99 100 101
6.8 0 95.7 (IC50 20 mM) 96.5 (IC50 37 mM) 100.2 (IC50 2.5 mM) 89.4 (IC50 10 mM)
0.490 0.468 0.859 0.767 0.406 0.052
E E A A B C
78.9 (IC50 45 mM) 43.2c (IC50 240 mM)
0.385 0.772
B D
P. tschimganica
R. maximowiczii
Cysteine a-Tocopherol
D E
G G F F G B F F C E
Standard deviations of values of these data were less than 1%. a The concentrations of compounds were 100 mM except for that of a-tocopherol. b Radical scavenging, shown as absorbance change at 517 nm due to scavenging of DPPH radicals by the compounds. Concentration, 20 mM. c Concentration, 200 mM.
ARTICLE IN PRESS K. Kogure et al. / Phytomedicine 11 (2004) 645–651
scavenge DPPH radical appreciably, although they strongly inhibited lipid peroxidation. To determine how those compounds that did not scavenge DPPH radical appreciably acted, we examined the effect of KT133 on lipid peroxidation of EyPC liposomes induced by the hydrophilic radical generator 2,20 -azobis(2-amidinopropane)hydrochloride (AAPH). Although KT133 strongly inhibited ADP/Fe2+-induced lipid peroxidation, it did not inhibit AAPH-induced lipid peroxidation (data not shown).
Discussion In this study, to find candidates among seeds from traditional crude drugs of Uzbekistan, we examined the effects of 48 compounds (chemical structures shown in Fig. 1) isolated from F. pallida, F. penninervis, I. macrophylla, P. pabularia, P. tschimganica and R. maximowiczii collected in Uzbekistan on ADP/Fe2+induced lipid peroxidation and the absorbance of DPPH radicals. The mechanism of ADP/Fe2+-induced lipid peroxidation was reported previously; i.e., reaction of chelated Fe2+ with ADP with endogenous hydroperoxide of EyPC at the membrane surface induces lipid peroxidation (Fukuzawa et al., 1995). For arrangement of the results, we plotted the values of inhibition percentages of lipid peroxidation against the absorbance change of DPPH radicals, and classified them into seven categories as follows: (A) potent inhibition and high radical scavenging, (B) potent inhibition and medium radical scavenging, (C) potent inhibition and low radical scavenging, (D)
Fig. 2. Relationship between inhibition of lipid peroxidation and absorbance change of DPPH radical by the compounds isolated from Ferula, Inula, Prangos and Rheum plants. Plots were classified into seven categories based on the intensities of inhibition and the extents of absorbance change. Compounds with similar basic structures are shown by the same symbol.
649
medium inhibition and high radical scavenging, (E) low inhibition and medium radical scavenging, (F) medium inhibition and no radical scavenging, and (G) low inhibition and no radical scavenging (see Fig. 2). Interestingly, there is no compound showing medium inhibition and medium radical scavenging. The mechanisms of action of the compounds were inferred from this classification. The inhibition of lipid peroxidation by compounds categorized as A, B and D should be due to their radicalscavenging ability. For example, the potent inhibitory effect of KT100 should be due to its radical-scavenging ability. As KT100 is a polyphenolic compound, and resembles the antioxidants reported previously (Matsuda et al., 2001), its phenolic-OH groups should be the scavenging site of radicals. The radical scavenging by the phenolic-OH groups in the galloyl moiety should also be responsible for the potent inhibitory effects of KT98 and KT99. The inhibitory effects of compounds in categories D and E were lower than those of the compounds in categories A, B and C, although these compounds have radical-scavenging ability. We suppose that as the interaction of compounds in categories D and E with the liposomal membrane are difficult, they did not prevent progress of peroxidation at the inside of the lipid membrane. Similarly, the absence of inhibitory effects of compounds such as KT97, having radicalscavenging ability, may be due to lack of interaction with the membrane. Of the compounds in category E, we paid attention to KT96. The structure of KT96 resembles that of KT97. KT97 has a phenolic OH group like general antioxidants (Noguchi and Niki, 2000), and showed significant radical scavenging ability. Interestingly, KT 96, which has no phenolic-OH, also showed radical scavenging ability. Perhaps a hydrogen atom at the carbon of the benzyl position of KT96 is pulled out by a free radical like cepharanthine as reported previously (Kogure et al., 1999). Similarly, KT110 also does not have a phenolic-OH, but scavenged DPPH radicals significantly. As shown in the Fig. 1, the structure of KT110 resembles those of KT118 and KT119, which have slight radical scavenging ability, except for the stereo structure of hydrogen at C5 and terminal short chain at C1’. Accordingly, these structural differences would be responsible for the radical scavenging ability of KT110. On the other hand, the inhibitory effect of compounds in categories C and F, which have low or no radical scavenging ability, were suggested to be due to another mechanism, such as chelation of Fe2+ or prevention of interaction of Fe2+ with the lipid membranes. To obtain information about the mechanism of the compounds in categories C and F, we examined the effects of KT133 on the absorbance spectra, with or without ADP/Fe2+.
ARTICLE IN PRESS 650
K. Kogure et al. / Phytomedicine 11 (2004) 645–651
The spectrum of KT133 was not affected by Fe2+, indicating that KT133 did not form a complex with Fe2+(data not shown). In addition, KT133 did not inhibit AAPH-induced lipid peroxidation. From these results, we suggest that inhibition by KT133 of ADP/ Fe2+-induced lipid peroxidation is due to prevention of interaction of Fe2+ with the liposomes possibly by some modification of the membrane, although the modification is unclear. We suppose that the inhibition mechanism of other compounds in categories C and E, such as KT118 and KT75, is almost the same as that of KT133. In addition, from comparison of KT133, KT129, KT131 and KT132, we suggest that the hydroxyl group in KT129 is responsible for radical scavenging, and terminal chloride is necessary for the potent inhibitory effect of KT133 on ADP/Fe2+-induced lipid peroxidation. In this study, from 48 compounds isolated from F. pallida, F. penninervis, I. macrophylla, P. pabularia, P. tschimganica and R. maximowiczii collected in Uzbekistan, we obtained various novel antioxidants. Nine especially potent antioxidants (KT98, 99, 100, 101, 108, 118, 119, 129 and 133) with various mechanisms of action were found. Of these, several compounds inhibited ADP/Fe2+-induced lipid peroxidation significantly, although radical-scavenging ability was lacking. Other compounds, which have no phenolic-OH groups, showed radical scavenging ability. More detailed investigations about each candidate are in progress to develop efficient antioxidative medicines.
References Aviram, M., 2000. Review of human studies on oxidative damage and antioxidant protection related to cardiovascular diseases. Free Radic. Res. 33, S85–S87. Cadenas, E., Davies, K.J., 2000. Mitochondrial free radical generation oxidative stress and aging. Free Radic. Biol. Med. 29, 222–230. Fukuzawa, K., Soumi, K., Iemura, M., Goto, S., Tokumura, A., 1995. Dynamics of xanthine oxidase- and Fe3+-ADPdependent lipid peroxidation in negatively charged phospholipid vesicles. Arch. Biochem. Biophys. 316, 83–91. Goto, S., Kogure, K., Abe, K., Kimata, Y., Kitahama, K., Yamashita, E., Terada, H., 2001. Efficient radical trapping at the surface and inside the phospholipid membrane is responsible for highly potent antiperoxidative activity of the carotenoid astaxanthin. Biochim. Biophys. Acta 1512, 251–258. Hsu, T.C., Young, M.R., Cmarik, J., Colburn, N.H., 2000. Activator protein 1 (AP-1)- and nuclear factor kappa B (NF-kappaB)-dependent transcriptional events in carcinogenesis. Free Radic. Biol. Med. 28, 1338–1348. Kogure, K., Sassa, H., Abe, K., Kitahama, K., Sano, Y., Kawano, H., Nakagawa, Y., Terada, H., 1998. Inhibitory effects of pentamethine trinuclear cyanine dyes on ADP/ Fe2+-induced lipid peroxidation in rat liver mitochondria:
changes in the mode of action with the hydrophobic nature of the dyes. Biol. Pharm. Bull. 21, 180–183. Kogure, K., Goto, S., Abe, K., Ohiwa, C., Akasu, M., Terada, H., 1999. Potent antiperoxidation activity of the bisbenzylisoquinoline alkaloid cepharanthine: the amine moiety is responsible for its pH-dependent radical scavenge activity. Biochim. Biophys. Acta 1426, 133–142. Kogure, K., Goto, S., Nishimura, M., Yasumoto, M., Abe, K., Ohiwa, C., Sassa, H., Kusumi, T., Terada, H., 2002. Mechanism of potent antioxidative effect of capsaicin. Biochim. Biophys. Acta 1573, 84–92. Matsuda, H., Morikawa, T., Toguchida, A., Park, J.Y., Harima, S., Yoshikawa, M., 2001. Antioxidant constituents from Rhubarb: structural requirements of stilbenes for the activity and structures of two new anthraquinone glucosides. Bioorg. Med. Chem. 9, 41–50. Noguchi, N., Niki, E., 2000. Phenolic antioxidants: a rationale for design and evaluation of a novel antioxidant drug for atherosclerosis. Free Radic. Biol. Med. 28, 1538–1546. Sassa, H., Takaishi, Y., Terada, H., 1990. Triterpene celastrol as a very potent inhibitor of lipid peroxidation in mitochondria. Biochem. Biophys. Res. Commun. 172, 890–897. Sassa, H., Kogure, K., Takaishi, Y., Terada, H., 1994. Structural basis of potent antiperoxidation activity of the triterpene celastrol in mitochondria: effect of negative membrane surface charge on lipid peroxidation. Free Radic. Biol. Med. 17, 201–207. Shikishima, Y., Takaishi, Y., Honda, G., Ito, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., 2001a. Phenylbutanoids and stilbene derivatives of Rheum maximowiczii. Phytochemistry 56, 377–381. Shikishima, Y., Takaishi, Y., Honda, G., Ito, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., 2001b. Terpenoids and g-pyrone derivatives from Prangos tshimganica. Phytochemistry 57, 135–141. Shikishima, Y., Takaishi, Y., Honda, G., Ito, M., Takada, Y., Kodzhimatov, O.K., Ashurmetov, O., Lee, K.-H., 2001c. Chemical constituents of Prangos tschimganica; structure elucidation and absolute configuration of coumarin and furanocoumarin derivatives with anti-HIV activity. Chem. Pharm. Bull. 49, 877–880. Shikishima, Y., Takaishi, T., Honda, G., Ito, M., Takeda, Y., Tori, M., Takaoka, S., Kodzhimatov, O.K., Ashurmetov, O., 2002. Sesquiterpenes from Ferula penninervis. J. Nat. Prod. 65, 1897–1903. Simonini, G., Pingnone, A., Generini, S., Falcini, F., Cerinic, M.M., 2000. Emerging potentials for an antioxidant therapy as a new approach to the treatment of systemic sclerosis. Toxicology 155, 1–15. Su, B.-N., Takaishi, Y., Yabuuchi, T., Kusumi, T., Tori, M., Takaoka, S., 2000a. The structures and absolute configurations of macrophylic acids A-E five new rearranged endesmane sesquiterpene acids from the bark of Inula macrophylla. Tetrahedron Lett. 41, 2395–2400. Su, B.-N., Takaishi, Y., Honda, G., Itoh, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., 2000b. Sesquiterpene phenylpropanoid and sesquiterpene chromone derivatives from Ferula pallida. J. Nat. Prod. 63, 520–522.
ARTICLE IN PRESS K. Kogure et al. / Phytomedicine 11 (2004) 645–651
Su, B.-N., Takaishi, Y., Yabuuchi, T., Kusumi, T., Tori, M., Takaoka, S., Honda, G., Ito, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., 2001. Sesquiterpenes and monoterpenes from the bark of Inula macrophylla. J. Nat. Prod. 64, 466–471. Tada, Y., Shikishima, Y., Takaishi, Y., Shibata, H., Higuti, T., Honda, G., Ito, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., Ohmoto, Y., 2002. Coumarins and gpyrone derivatives from Prangos pabularia: antibacterial activity and inhibition of cytokine release. Phytochemistry 59, 649–654.
651
Tamemoto, K., Takaishi, Y., Chen, B., Kawazoe, K., Shibata, H., Higuti, T., Honda, G., Ito, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., 2001. Sesquiterpenoids from the fruits of Ferula kuhisutanica and antibacterial activity of the constituents of F. kuhisutanaica. Phytochemistry 58, 763–767. Zhou, P., Takaishi, Y., Duan, H., Chen, B., Honda, G., Ito, M., Takada, Y., Kodzhimatov, K.O., Lee, K.-H., 2000. Coumarins and biocoumarin from Ferula sumbul: anti-HIV activity and inhibition of cytokine release. Phytochemistry 53, 689–697.
Phytomedicine 11 (2004) 645–651 www.elsevier.de/phymed
Novel antioxidants isolated from plants of the genera Ferula, Inula, Prangos and Rheum collected in Uzbekistan K. Kogure, I. Yamauchi, A. Tokumura, K. Kondou, N. Tanaka, Y. Takaishi, K. Fukuzawa Faculty of Pharmaceutical Sciences, University of Tokushima, Shomachi-1, 770-8505 Tokushima, Japan Received 7 March 2003; accepted 5 September 2003
Abstract We examined the effects of 48 compounds isolated from Ferula pallida, F. penninervis, Inula macrophylla, Prangos pabularia, P. tschimganica and Rheum maximowiczii collected in Uzbekistan on ADP/Fe2+-induced lipid peroxidation of egg yolk phosphatidylcholine liposomes. Of those compounds, 23 inhibited ADP/Fe2+-induced lipid peroxidation and nine showed especially strong inhibition of lipid peroxidation. Most compounds that inhibited peroxidation scavenged the 1,10 -diphenyl-2-picrylhydrazyl (DPPH) radical, indicating that the inhibition was due to radical scavenging. However, some compounds did not scavenge DPPH but inhibited lipid peroxidation significantly, suggesting that their inhibitory effect was not due to radical scavenging but to some other mechanism, such as prevention of Fe2+ function. Thus, we found various new antioxidants, some of which had a unique mechanism of action, in Ferula, Inula, Prangos and Rheum plants collected in Uzbekistan as seeds used in medicine. r 2004 Elsevier GmbH. All rights reserved. Keywords: Antioxidants; Natural products; Ferula; Inula; Prangos; Rheum
Introduction Oxidative stresses are suggested to be related to various diseases and pathological conditions such as atherosclerosis (Aviram, 2000), carcinogenesis (Hsu et al., 2000) and aging (Cadenas and Davies, 2000). Thus, to eliminate oxidative stresses such as reactive oxygen species in the body, many workers have been attemptings to develop antioxidative drugs (Simonini et al., 2000). We have studied the antioxidation effects of various compounds, such as the triterpene celastrol (Sassa et al., 1990, 1994), the bisbenzylisoquinoline alkaloid cepharanthine (Kogure et al., 1999), the Corresponding author. Tel.: +81-88-633-7248; fax:+81-88-6339572. E-mail address: [email protected] (K. Fukuzawa).
0944-7113/$ - see front matter r 2004 Elsevier GmbH. All rights reserved. doi:10.1016/j.phymed.2003.09.004
xanthophyll astaxanthine (Goto et al., 2001), the pungent ingredient of red pepper known as capsaicin (Kogure et al., 2002) and artificial pigment cyanine dyes (Kogure et al., 1998), in order to develop antioxidants for medical use. Recently, we examined traditional crude drugs of central Asian people, and found various seed compounds (Su et al., 2000a, b, 2001; Zhou et al., 2000; Shikishima et al., 2001a-c; Tamemoto et al., 2001; Tada et al., 2002; Shikishima et al., in press). As traditional crude drugs in Uzbekistan have never been examined scientifically, the compounds in these crude drugs are very likely to contain new medicinal compounds. Recently, we reported that compounds isolated from various plants collected in Uzbekistan showed anti-HIV activity in vitro (Shikishima et al., 2001c) and preventive effects on the generation and release of inflammation
ARTICLE IN PRESS 646
K. Kogure et al. / Phytomedicine 11 (2004) 645–651
agents such as TNF-a and IL-2 in vitro (Tada et al., 2002; Shikishima et al., 2002). These biological effects might thus be associated with antioxidative activity. In this study, to find candidates among medicinal seeds from traditional Uzbek crude drugs, we examined the effects of 48 compounds isolated from Ferula pallida, F. penninervis, Inula macrophylla, Prangos pabularia, P. tschimganica and Rheum maximowiczii collected in Uzbekistan on ADP/Fe2+-induced lipid peroxidation of egg yolk phosphatidylcholine liposomes (Fukuzawa et al., 1995).
Assay of radical scavenging The radical scavenging activities of test compounds (20 mM) in the ethanolic solution were determined from the decrease in the absorbance of DPPH radical at 517 nm due to their scavenging of an unpaired electron of the stable DPPH radical in a mixture of 10 ml of ethanol, 10 ml of 50 mM 2-(N-morpholino)ethanesulfonic acid (MES) buffer (pH 5.5) and 5 ml of 0.5 mM DPPH in ethanol (Sassa et al., 1990, 1994; Kogure et al., 1999).
Measurements of absorption spectra
Materials and methods Materials The plants F. pallida, F. penninervis, I. macrophylla, P. pabularia, P. tschimganica and R. maximowiczii were collected in Uzbekistan, and many natural compounds were isolated from them as described previously (Su et al., 2000a b, 2001; Zhou et al., 2000; Shikishima et al., 2001a–c; Tamemoto et al., 2001; Tada et al., 2002; Shikishima et al., in press). Egg yolk phosphatidylcholine (EyPC) was purchased from Nihon Seika Co. (Tokyo). 1,10 -Diphenyl-2-picrylhydrazyl (DPPH) was obtained from Nacalai Tesque (Kyoto). a-Tocopherol was kindly provided by Eisai Co. (Tokyo, Japan). Other reagents were of the highest grade commercially available. The solution of Fe2+ was prepared by dissolving FeSO4 in water just before use, to avoid precipitation of Fe3+ as Fe(OH)3.
Assay of lipid peroxidation Liposomes consisting of EyPC were prepared by sonication in a bath-type sonicator in 10 mM Tris-HCl buffer (pH 7.4). Solutions of the compounds with dimethyl sulfoxide were added to liposomes (0.2 mmol lipid/ml) suspended with 10 mM Tris-HCl buffer (pH 7.4) at 25 1C. Immediately after their addition, 1 mM ADP and 0.1 mM FeSO4 were added to the suspension to induce lipid peroxidation. After 15 min, an ethanolic solution of the well-known antioxidant butylated hydroxytoluene (BHT) was added to the reaction mixture (final concentration, 4.5 mM) to terminate lipid peroxidation. Amounts of lipid peroxides were determined as amounts of thiobarbituric acid-reactive substances (TBARS) in terms of malonedialdehyde (MDA), using tetraethoxypropane as a standard (Goto et al., 2001).
The absorption spectra of test samples in 10 mM TrisHCl buffer (pH 7.4) were measured on a spectrophotometer (Shimadzu UV-1600) at 25 1C.
Results We examined the effects of 48 compounds (chemical structures shown in Fig. 1) isolated from F. pallida, F. penninervis, I. macrophylla, P. pabularia, P. tschimganica and R. maximowiczii collected in Uzbekistan on ADP/Fe2+-induced lipid peroxidation of EyPC liposomes, by comparison with those of the well-known antioxidants cysteine and a-tocopherol. Cysteine and atocopherol inhibited liposomal lipid peroxidation (Table 1). The inhibitory effect of cysteine was potent in this system. As listed in Table 1, 23 compounds (final concentrations, 100 mM) inhibited lipid peroxidation. The inhibition percentages of nine compounds (KT98, 99, 100, 101, 108, 118, 119, 129 and 133) were over 70% as compared to that of cysteine. KT100 and KT118 were required at especially low concentrations for 50% inhibition (IC50 values) (2.5 mM and 0.7 mM, respectively). Among the compounds isolated from F. pallida and P. pabularia, no potent antioxidant was observed. The number of potent antioxidative compounds in R. maximowiczii was the highest. To obtain information about the mechanisms of the antioxidative effects of the compounds, we examined their radical scavenging effects by measuring absorbance change of DPPH radical at 517 nm (Table 1). We used cysteine as a standard, one molecule of which scavenges one DPPH radical (Sassa et al., 1990). In our system, addition of cysteine (final concentration, 20 mM) immediately decreased absorbance at 517 nm by scavenging DPPH radical. Over 2/3 of the compounds that inhibited lipid peroxidation scavenged DPPH radical. The compounds that scavenged most strongly were KT08, 110, 96, 97, 98, 99 and 100. However, some compounds (KT101, 108, 118, 119 and 133) did not
ARTICLE IN PRESS K. Kogure et al. / Phytomedicine 11 (2004) 645–651
647
Fig. 1. Chemical structures of cysteine, a-tocopherol and compounds isolated from Ferula, Inula, Prangos and Rheum plants.
ARTICLE IN PRESS 648
K. Kogure et al. / Phytomedicine 11 (2004) 645–651
Table 1. Values of inhibition percentage of lipid peroxidation and absorbance change due to scavenging of DPPH radical by compounds isolated from Ferula, Inula, Prangos and Rheum plants, and categories of them classified in Fig. 2 Plant
Compound
Inhibition of lipid peroxidation(%)a
Radical scavenging (DA517 )b
Category
F. pallida
KT KT KT KT KT KT KT KT
23 29 31 34 35 36 37 38
16.4 7.7 0 0 0 0 0 0
0.158 0 0 — — — — —
E G G
F. penninervis
KT KT KT KT KT KT KT KT KT KT KT KT KT
103 104 106 107 108 110 111 116 117 118 119 120 121
0 0 0 0 83.9 (IC50 60 mM) 15.7 0 0 0 103.0 (IC50 0.7 mM) 72.1 (IC50 70 mM) 0 0
— — — — 0.004 0.634 — — — 0.130 0.143 — —
C E
C C
I. macrophylla
KT 08 KT 10 KT 11
35.9 5.8 0
0.648 0.232 —
P. pabularia
KT KT KT KT KT KT KT
64 66 68 71 72 74 75
0 0 0 0 4.3 11.8 40.6
— — — — 0.004 0.043 0.069
KT KT KT KT KT KT KT KT KT
123 125 127 129 130 131 132 133 135
28.8 0 2.1 91.7 (IC50 50 mM) 25.7 18.6 0 87.2 (IC50 75 mM) 8.0
0 — 0 0.331 0.044 0 — 0.019 0.082
KT KT KT KT KT KT
96 97 98 99 100 101
6.8 0 95.7 (IC50 20 mM) 96.5 (IC50 37 mM) 100.2 (IC50 2.5 mM) 89.4 (IC50 10 mM)
0.490 0.468 0.859 0.767 0.406 0.052
E E A A B C
78.9 (IC50 45 mM) 43.2c (IC50 240 mM)
0.385 0.772
B D
P. tschimganica
R. maximowiczii
Cysteine a-Tocopherol
D E
G G F F G B F F C E
Standard deviations of values of these data were less than 1%. a The concentrations of compounds were 100 mM except for that of a-tocopherol. b Radical scavenging, shown as absorbance change at 517 nm due to scavenging of DPPH radicals by the compounds. Concentration, 20 mM. c Concentration, 200 mM.
ARTICLE IN PRESS K. Kogure et al. / Phytomedicine 11 (2004) 645–651
scavenge DPPH radical appreciably, although they strongly inhibited lipid peroxidation. To determine how those compounds that did not scavenge DPPH radical appreciably acted, we examined the effect of KT133 on lipid peroxidation of EyPC liposomes induced by the hydrophilic radical generator 2,20 -azobis(2-amidinopropane)hydrochloride (AAPH). Although KT133 strongly inhibited ADP/Fe2+-induced lipid peroxidation, it did not inhibit AAPH-induced lipid peroxidation (data not shown).
Discussion In this study, to find candidates among seeds from traditional crude drugs of Uzbekistan, we examined the effects of 48 compounds (chemical structures shown in Fig. 1) isolated from F. pallida, F. penninervis, I. macrophylla, P. pabularia, P. tschimganica and R. maximowiczii collected in Uzbekistan on ADP/Fe2+induced lipid peroxidation and the absorbance of DPPH radicals. The mechanism of ADP/Fe2+-induced lipid peroxidation was reported previously; i.e., reaction of chelated Fe2+ with ADP with endogenous hydroperoxide of EyPC at the membrane surface induces lipid peroxidation (Fukuzawa et al., 1995). For arrangement of the results, we plotted the values of inhibition percentages of lipid peroxidation against the absorbance change of DPPH radicals, and classified them into seven categories as follows: (A) potent inhibition and high radical scavenging, (B) potent inhibition and medium radical scavenging, (C) potent inhibition and low radical scavenging, (D)
Fig. 2. Relationship between inhibition of lipid peroxidation and absorbance change of DPPH radical by the compounds isolated from Ferula, Inula, Prangos and Rheum plants. Plots were classified into seven categories based on the intensities of inhibition and the extents of absorbance change. Compounds with similar basic structures are shown by the same symbol.
649
medium inhibition and high radical scavenging, (E) low inhibition and medium radical scavenging, (F) medium inhibition and no radical scavenging, and (G) low inhibition and no radical scavenging (see Fig. 2). Interestingly, there is no compound showing medium inhibition and medium radical scavenging. The mechanisms of action of the compounds were inferred from this classification. The inhibition of lipid peroxidation by compounds categorized as A, B and D should be due to their radicalscavenging ability. For example, the potent inhibitory effect of KT100 should be due to its radical-scavenging ability. As KT100 is a polyphenolic compound, and resembles the antioxidants reported previously (Matsuda et al., 2001), its phenolic-OH groups should be the scavenging site of radicals. The radical scavenging by the phenolic-OH groups in the galloyl moiety should also be responsible for the potent inhibitory effects of KT98 and KT99. The inhibitory effects of compounds in categories D and E were lower than those of the compounds in categories A, B and C, although these compounds have radical-scavenging ability. We suppose that as the interaction of compounds in categories D and E with the liposomal membrane are difficult, they did not prevent progress of peroxidation at the inside of the lipid membrane. Similarly, the absence of inhibitory effects of compounds such as KT97, having radicalscavenging ability, may be due to lack of interaction with the membrane. Of the compounds in category E, we paid attention to KT96. The structure of KT96 resembles that of KT97. KT97 has a phenolic OH group like general antioxidants (Noguchi and Niki, 2000), and showed significant radical scavenging ability. Interestingly, KT 96, which has no phenolic-OH, also showed radical scavenging ability. Perhaps a hydrogen atom at the carbon of the benzyl position of KT96 is pulled out by a free radical like cepharanthine as reported previously (Kogure et al., 1999). Similarly, KT110 also does not have a phenolic-OH, but scavenged DPPH radicals significantly. As shown in the Fig. 1, the structure of KT110 resembles those of KT118 and KT119, which have slight radical scavenging ability, except for the stereo structure of hydrogen at C5 and terminal short chain at C1’. Accordingly, these structural differences would be responsible for the radical scavenging ability of KT110. On the other hand, the inhibitory effect of compounds in categories C and F, which have low or no radical scavenging ability, were suggested to be due to another mechanism, such as chelation of Fe2+ or prevention of interaction of Fe2+ with the lipid membranes. To obtain information about the mechanism of the compounds in categories C and F, we examined the effects of KT133 on the absorbance spectra, with or without ADP/Fe2+.
ARTICLE IN PRESS 650
K. Kogure et al. / Phytomedicine 11 (2004) 645–651
The spectrum of KT133 was not affected by Fe2+, indicating that KT133 did not form a complex with Fe2+(data not shown). In addition, KT133 did not inhibit AAPH-induced lipid peroxidation. From these results, we suggest that inhibition by KT133 of ADP/ Fe2+-induced lipid peroxidation is due to prevention of interaction of Fe2+ with the liposomes possibly by some modification of the membrane, although the modification is unclear. We suppose that the inhibition mechanism of other compounds in categories C and E, such as KT118 and KT75, is almost the same as that of KT133. In addition, from comparison of KT133, KT129, KT131 and KT132, we suggest that the hydroxyl group in KT129 is responsible for radical scavenging, and terminal chloride is necessary for the potent inhibitory effect of KT133 on ADP/Fe2+-induced lipid peroxidation. In this study, from 48 compounds isolated from F. pallida, F. penninervis, I. macrophylla, P. pabularia, P. tschimganica and R. maximowiczii collected in Uzbekistan, we obtained various novel antioxidants. Nine especially potent antioxidants (KT98, 99, 100, 101, 108, 118, 119, 129 and 133) with various mechanisms of action were found. Of these, several compounds inhibited ADP/Fe2+-induced lipid peroxidation significantly, although radical-scavenging ability was lacking. Other compounds, which have no phenolic-OH groups, showed radical scavenging ability. More detailed investigations about each candidate are in progress to develop efficient antioxidative medicines.
References Aviram, M., 2000. Review of human studies on oxidative damage and antioxidant protection related to cardiovascular diseases. Free Radic. Res. 33, S85–S87. Cadenas, E., Davies, K.J., 2000. Mitochondrial free radical generation oxidative stress and aging. Free Radic. Biol. Med. 29, 222–230. Fukuzawa, K., Soumi, K., Iemura, M., Goto, S., Tokumura, A., 1995. Dynamics of xanthine oxidase- and Fe3+-ADPdependent lipid peroxidation in negatively charged phospholipid vesicles. Arch. Biochem. Biophys. 316, 83–91. Goto, S., Kogure, K., Abe, K., Kimata, Y., Kitahama, K., Yamashita, E., Terada, H., 2001. Efficient radical trapping at the surface and inside the phospholipid membrane is responsible for highly potent antiperoxidative activity of the carotenoid astaxanthin. Biochim. Biophys. Acta 1512, 251–258. Hsu, T.C., Young, M.R., Cmarik, J., Colburn, N.H., 2000. Activator protein 1 (AP-1)- and nuclear factor kappa B (NF-kappaB)-dependent transcriptional events in carcinogenesis. Free Radic. Biol. Med. 28, 1338–1348. Kogure, K., Sassa, H., Abe, K., Kitahama, K., Sano, Y., Kawano, H., Nakagawa, Y., Terada, H., 1998. Inhibitory effects of pentamethine trinuclear cyanine dyes on ADP/ Fe2+-induced lipid peroxidation in rat liver mitochondria:
changes in the mode of action with the hydrophobic nature of the dyes. Biol. Pharm. Bull. 21, 180–183. Kogure, K., Goto, S., Abe, K., Ohiwa, C., Akasu, M., Terada, H., 1999. Potent antiperoxidation activity of the bisbenzylisoquinoline alkaloid cepharanthine: the amine moiety is responsible for its pH-dependent radical scavenge activity. Biochim. Biophys. Acta 1426, 133–142. Kogure, K., Goto, S., Nishimura, M., Yasumoto, M., Abe, K., Ohiwa, C., Sassa, H., Kusumi, T., Terada, H., 2002. Mechanism of potent antioxidative effect of capsaicin. Biochim. Biophys. Acta 1573, 84–92. Matsuda, H., Morikawa, T., Toguchida, A., Park, J.Y., Harima, S., Yoshikawa, M., 2001. Antioxidant constituents from Rhubarb: structural requirements of stilbenes for the activity and structures of two new anthraquinone glucosides. Bioorg. Med. Chem. 9, 41–50. Noguchi, N., Niki, E., 2000. Phenolic antioxidants: a rationale for design and evaluation of a novel antioxidant drug for atherosclerosis. Free Radic. Biol. Med. 28, 1538–1546. Sassa, H., Takaishi, Y., Terada, H., 1990. Triterpene celastrol as a very potent inhibitor of lipid peroxidation in mitochondria. Biochem. Biophys. Res. Commun. 172, 890–897. Sassa, H., Kogure, K., Takaishi, Y., Terada, H., 1994. Structural basis of potent antiperoxidation activity of the triterpene celastrol in mitochondria: effect of negative membrane surface charge on lipid peroxidation. Free Radic. Biol. Med. 17, 201–207. Shikishima, Y., Takaishi, Y., Honda, G., Ito, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., 2001a. Phenylbutanoids and stilbene derivatives of Rheum maximowiczii. Phytochemistry 56, 377–381. Shikishima, Y., Takaishi, Y., Honda, G., Ito, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., 2001b. Terpenoids and g-pyrone derivatives from Prangos tshimganica. Phytochemistry 57, 135–141. Shikishima, Y., Takaishi, Y., Honda, G., Ito, M., Takada, Y., Kodzhimatov, O.K., Ashurmetov, O., Lee, K.-H., 2001c. Chemical constituents of Prangos tschimganica; structure elucidation and absolute configuration of coumarin and furanocoumarin derivatives with anti-HIV activity. Chem. Pharm. Bull. 49, 877–880. Shikishima, Y., Takaishi, T., Honda, G., Ito, M., Takeda, Y., Tori, M., Takaoka, S., Kodzhimatov, O.K., Ashurmetov, O., 2002. Sesquiterpenes from Ferula penninervis. J. Nat. Prod. 65, 1897–1903. Simonini, G., Pingnone, A., Generini, S., Falcini, F., Cerinic, M.M., 2000. Emerging potentials for an antioxidant therapy as a new approach to the treatment of systemic sclerosis. Toxicology 155, 1–15. Su, B.-N., Takaishi, Y., Yabuuchi, T., Kusumi, T., Tori, M., Takaoka, S., 2000a. The structures and absolute configurations of macrophylic acids A-E five new rearranged endesmane sesquiterpene acids from the bark of Inula macrophylla. Tetrahedron Lett. 41, 2395–2400. Su, B.-N., Takaishi, Y., Honda, G., Itoh, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., 2000b. Sesquiterpene phenylpropanoid and sesquiterpene chromone derivatives from Ferula pallida. J. Nat. Prod. 63, 520–522.
ARTICLE IN PRESS K. Kogure et al. / Phytomedicine 11 (2004) 645–651
Su, B.-N., Takaishi, Y., Yabuuchi, T., Kusumi, T., Tori, M., Takaoka, S., Honda, G., Ito, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., 2001. Sesquiterpenes and monoterpenes from the bark of Inula macrophylla. J. Nat. Prod. 64, 466–471. Tada, Y., Shikishima, Y., Takaishi, Y., Shibata, H., Higuti, T., Honda, G., Ito, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., Ohmoto, Y., 2002. Coumarins and gpyrone derivatives from Prangos pabularia: antibacterial activity and inhibition of cytokine release. Phytochemistry 59, 649–654.
651
Tamemoto, K., Takaishi, Y., Chen, B., Kawazoe, K., Shibata, H., Higuti, T., Honda, G., Ito, M., Takeda, Y., Kodzhimatov, O.K., Ashurmetov, O., 2001. Sesquiterpenoids from the fruits of Ferula kuhisutanica and antibacterial activity of the constituents of F. kuhisutanaica. Phytochemistry 58, 763–767. Zhou, P., Takaishi, Y., Duan, H., Chen, B., Honda, G., Ito, M., Takada, Y., Kodzhimatov, K.O., Lee, K.-H., 2000. Coumarins and biocoumarin from Ferula sumbul: anti-HIV activity and inhibition of cytokine release. Phytochemistry 53, 689–697.