Kobe University Repository : Thesis

学位論文題目Tit le

γ-PYRONYL_TRITERPENOID SAPONIN(CHROMOSAPONIN Ｉ) ASAN ELECTRON DONOR(γ-ピロニルートリテルペノイドサポニン(クロモサポニンⅠ)の電子供与性)

氏名Author Tsujino, Yoshio

専攻分野Degree 博士（理学）

学位授与の日付Date of Degree 1994-09-30

資源タイプResource Type Thesis or Dissertat ion / 学位論文

報告番号Report Number 甲1329

権利Rights

JaLCDOI 10.11501/3105351

URL http://www.lib.kobe-u.ac.jp/handle_kernel/D1001329※当コンテンツは神戸大学の学術成果です。無断複製・不正使用等を禁じます。著作権法で認められている範囲内で、適切にご利用ください。

Create Date: 2018-06-18

Doctoral Dissertation

y-PYRONYL-TRITERPENOID SAPONIN

(CHROMOSAPONIN I)

AS AN ELECTRON DONOR

by

Y oshio Tsujino

THE GRADUATE SCHOOL OF SCIENCE AND TECHNOLOGY

KOBE UNIVERSITY

August 1994

Doctoral Dissertation

y-PYRONYL-TRITERPENOID SAPONIN

(CHROMOSAPONIN I)

AS AN ELECTRON DONOR

y- ~ 0 .:: )v - r 1) T )V ~ / l' r ~ *' .:: / (7 D:C~ *'.:: / I) 0)

-m: -r1~!} tt

by

Yoshio Tsujino

THE GRADUATE SCHOOL OF SCIENCE AND TECHNOLOGY

KOBE UNIVERSITY

August 1994

Doctoral Dissertation

γPYRONYL-1、RITERPENOIDSAPONIN

(CHROMOSAPONIN 1)

AS AN ELECTRON DONOR

子ピロニルートリテルペノイドサボニン

(クロモサボニン 1)の

電子供与性

by

Yoshio Tsujino

THE GRADUATE SCHOOL OF SCIENCE AND TECHNOLOGY

KOBE UNIVERSITY

August 1994

TABLE OF CONTENTS

List of Schemes, Tables, and Figures .... · .. ································ ........ ·.····· .... 1

Summary ...................................................... ············································3

Abbreviations········ ................................................................................ ····5

General Introduction···············································································.·. 7

Chapter 1

Electrochemical and Spectrophotometrical Studies on CSI

I. Introduction·························································································l 0

II. Materials and Methods········································································· 10

II -1. Preparation of CSI, SI and Cyt c ..................................................... 10

11-2. Chemicals················································································ ·11

11-3. Electrochemical measurements with a GC electrode and a PFC electrode ········11

II -4. Spectrophotometric measurements····· ............................................... 12

11-5. Other measurements (HPLC and O2 uptake)· ................................ ········13

III. Results and Discussion······································································· 13

III -1. Electrochemical behavior of CSI at the GC electrode ............................... 13

111-2. Coulometric measurement ofCSI ·····················································19

III -3. Voltammetric comparison of the oxidation potentials of CSI and biological

antioxidants· .............................................................................. 1 9

III -4. Reductions of Quinones and Cyt c by CSI ........................................... 20

111-5. Kinetic analysis of the reduction ofBQ by CSI······································26

III -6. pH Dependence of kapp for CSI and the comparison with those for other

reductants ................................................................................. 3 1

III -7. pH Dependence of the Epa value for CSI ············································33

111-8. Electrochemical behaviors of Quinones at the PFC electrode ·······················33

Chapter 2

Antioxidative Effects of CSI

I. Introduction························································································ ·40

II. Materials and Methods········································································· 41

II -1. Preparation of PC· ....................................................................... 41

11-2. Chemicals··· ............................................. '" ....................... ·······41

II -3. Reaction with a stable free radical, galvinoxyl, in EtOH· .......................... ·41

11-4. Reaction with a stable free radical, DPPH, in aqueous EtOH .............. ········42

11-5. Reaction with a radical initiator, AAPH, in aqueous solution····················· ·43

11-6. Inhibition of the formation ofPC-OOH in the AAPH system······················43

II -7. Determination of O2 - with NBT and CLA ........................................... ·44

III. Results and Discussion······································································· 45

III -1. Reaction with a stable free radical, galvinoxyl, in EtOH· .......................... ·45

111-2. Reaction with a stable free radical, DPPH, in aqueous EtOH· ........ ·············47

III -3. Reaction with a radical from AAPH in aqueous solution· .......................... 51

111-4. Antioxidative activity in the PC-AAPH system··· ................................. ··53

111-5. O2- generation by CSI under aerobic condition······································ 55

Chapter 3

Implications and Perspectives

I. Characteristics of CSI as an Electron Donor ........................................... ·61

II. A Possibility of Participation of CSI as a pH Sensor in the Electron

Transfer Induced by pH Change in vivo .......................................... 62

III. Antioxidative Effects of CSI and a Proposed Cyclic Mechanism .......... · .. ·64

References ...................................................... ·······································70

Publication List ...................................................... ·································76

Acknowledgments ...................................................... ······························78

ii

List of Schemes, Tables, and Figures

Scheme I. Mediator function ofBQ in the reduction ofCyt c by CSI ························23

Scheme II. Proton/electron transfer diagrams for CSI and ascorbate ...................... · .... 65

Scheme III. Proposed cyclic mechanism for the radical-scavenging reaction by CSI

in living cells·· ................................................................... ·······67

Table I. Anodic peak potentials, Epa' of CSI and other biological antioxidants ··············20

Table II. Stoichiometry in the reduction of 100,UM DPPH by 10 ,UM reductants

(ascorbate, CSI, and cysteine) under aerobic and anaerobic conditions·············· 51

Fig. 1. Chemical structures of CSI, y-Pyr, and SI ·················································8

Fig. 2. Cyclic voltammograms of CSI and y-Pyr recorded with a GC electrode

in 0.1 M phosphate buffer (pH 7.7) ........................................................ 14

Fig. 3. Plot ofthe anodic peak current (Ipa) against the voltage scan rate (v)

for cyclic voltammetry ........................................................................ 16

Fig. 4. Effect of the CSI concentration on anodic peak current··································l 7

Fig. 5. Cyclic voltammograms showing the adsorption behavior of CSI

on the GC electrode surface· ................................................................. 18

Fig. 6. Spectral change for the BQ-CSI system· .................................................. ·21

Fig. 7. Reductions of BQ by ascorbate, CSI, Trolox, and urate ................................ ·22

Fig. 8. Reductions of quinones by CSI ............................................................. 24

Fig. 9. Reduction of Cyt c by CSI and its acceleration by BQ .................................. ·25

Fig. 10. Time-course of the reduction ofBQ by CSI ............................................ ·27

Fig. 11. Decrease of A245 ofBQ and plotting of the reciprocal of [BQ] against time

on the reduction ofBQ by CSI at various pH's ........................................ ·29

1

Fig. 12. pH Dependence of the apparent second-order rate constants (kapp)

for the reductions of BQ by CSI, NADPH, and Trolox,

and for the reduction of 2,5-DMBQ by ascorbate .. ································· .. ·· 32

Fig. 13. pH Dependence of cyclic voltammograms of CSI recorded with a GC electrode··· 34

Fig. 14. Cyclic voltammograms ofBQ recorded with a PFC electrode and a GC electrode

....................................................... ··········································35

Fig. 15. Effect of voltage scan rates (v) on cyclic voltammograms of BQ recorded

with a PFC electrode· ....................................................................... 36

Fig. 16. pH Dependence of cyclic voltammograms of BQ with a PFC electrode ·············37

Fig. 17. pH Dependence of the midpoint potentials (Emid) for quinones with a PFC electrode

...................................................... ···········································38

Fig. 18. Reactions of CSI, SI, y-Pyr, and a-Toc with a stable free radical,galvinoxyl ··46

Fig. 19. Reactions ofCSI, SI, y-Pyr, and other reductants with a stable free radical, DPPH

...................................................... ···········································48

Fig. 20. Effects of pH on the reduction of DPPH by CSI········································49

Fig. 21. O2 uptake in the DPPH-CSI system .... ···· ...... ··· ........ · ...... ·· .. · .. ··· .. ···· .. · .. ·50

Fig. 22. Reactions of CSI, SI, y-Pyr, and of other reductants

Fig. 23.

Fig. 24.

Fig. 25.

Fig. 26.

Fig. 27.

with radicals induced by AAPH· ......................................................... ·52

Antioxidative effects of CSI, SI, y-Pyr and other reductants (urate and Trolox)

on the PC-OOH formation in the oxidation of PC induced by AAPH ··············54

Spectral changes for the interaction of NBT with CSI and for the reduction of NBT

in the presence of CSI and AAPH· .... ···· .. · .... ···· .. ······· .. ······ .. ··················56

Effect of SOD on the formation of diformazon from NBT

in the presence of CSI and AAPH .. · .... · .. ········ .. · .. ····· .. ········· .. ······· .. ·······57

Increase of chemiluminescence from CLA by adding CSI and AAPH,

and the decrease of it by adding SOD· ................................................... ·58

Effect of CSI on the chemiluminescence from CLA induced

by HX-XOD system at 25°C .. ····· .. ··· .. ····················· .. ··· .. · .. ·· .. ····· .. ······59

2

Summary

The 'fpyronyl-triterpenoid saponin, named chromosaponin I (CSI), isolated from pea

(Pisum sativum L. cv Alaska), was studied electrochemically and biochemically to be

characterized as a new amphipathic reductant.

Cyclic voltammetric analysis demonstrated the existence of a definite reducing power,

stronger than cysteine and comparable to ascorbate and urate at pH 7.7. The 'tpyronyl moiety

is responsible for the electroactivity. CSI was strongly adsorbed at a glassy-carbon electrode to

form a monomolecular layer or multimolecular layers on the electrode surface, depending on its

concentration. From a comparison with other reductants, the formal redox potential (EO,) of

CSI was estimated to be 0.3-0.4 V (pH 7.0) vs. normal hydrogen electrode (NHE).

(Chapter 1)

Chromosaponin I reduced 1,4-benzoquinone (BQ), methyl-l,4-benzoquinone, and

cytochrome c, but did not effectively reduce 2,5-dimethyl-l,4-benzoquinone, indicating that

CSI reduces the compounds having the EO' value more positive than about 0.20 V (pH 7.0)

vs. NHE. As to the reaction with BQ, the reducing power of CSI is comparable to NADPH

and 2-carboxy-2,5,7,8-tetramethyl-6-chromanol (Trolox, a water-soluble analogue for a­

tocopherol) in the physiological pH. CSI reacts with BQ with a stoichiometry of 1 : 1, and two

electrons participate in the redox reaction. The apparent rate constants (kapp) follow the rate

law -d[A]/dt = kapp[A][CSI], where A is the acceptor BQ. The kapp for the CSI-quinone

system shows a strong pH dependence, the slope ~log kappl~pH (= 0.93) being about three

times larger than that for the ascorbate-quinone system. The strongly pH dependent reducing

power of CSI is further confirmed by means of cyclic voltammetry. It is also shown that CSI

can reduce cytochrome c at the physiological pH quite effectively in the presence of BQ as a

mediator. (Chapter 1)

3

Chromosaponin I scavenges a hydrophobic stable free radical, a,a-diphenyl-,B­

picrylhydrazyl in aqueous ethanol, more rapidly at alkaline pH. One mol of CSI reduced 1.1

mol of the radical under aerobic condition and 1.7 mol under anaerobic condition. CSI also

inhibited the oxidation of soybean phosphatidylcholine liposomal membranes induced by a

water-soluble radical initiator, 2,2' -azobis(2-amidinopropane) dihydrochloride. The antioxidant

activity of CSI is similar to that of urate for at least 60 min after mixing. Soyasaponin I, a

degraded product of CSI, which was previously thought to be an antioxidant exerted no

antioxidant activity. Although CSI radicals produced in the radical scavenging reactions

generated 02- under an air atmosphere, in living cells CSI is expected to be an effective radical

scavenger because of its concentrations (2-3 mM in the hook and the root tip) higher than that

of 02 (assumed to be 3-30 pM). (Chapter 2)

Biological reductants have been classified into two groups; hydrophobic (e.g.,

tocopherol) and hydrophilic (e.g., ascorbate). CSI is a new type of reductant, i.e.

amphipathic. A working hypothesis, that CSI is a biological pH-sensor and a radical scavenger

in living cells, was presented and discussed.

4

Abbreviations

A: dehydroascorbate

A2-: ascorbate dianion

A-·: semidehydroascorbate radical anion

AAPH (A -N =N -A): 2,2' -azobis(2-amidinopropane )dihydrochloride

AH2: ascorbic acid

AH-: ascorbate monoanion

AH·: semidehydroascorbate neutral anion

BQ: 1,4-benzoquinone

CLA: 2-methyl-6-phenyl-3, 7 -dihydroimidazo[ I ,2-a ]pyrezin-3-one

CSI: chromosaponin I

CSI-: anion form in the /'"pyronyl moiety of CSI molecules

CSI·: CSI neutral radical

CSI+·: CSI radical cation

CSI(Glut: anion form in the glucuronopyranosyl moiety of CSI molecules

Cyt c: cytochrome c

DPPH: a,a-diphenyl-{J-picrylhydrazyl

DF: diformazon of NBT

2,5-DMBQ: 2,5-dimethyl-I,4-benzoquinone

EtOH: ethanol

GC: glassy carbon

HOAc: acetic acid

HPLC: high performance liquid chromatography

HQ: hydroquinone (1,4-benzenediol)

HX: hypoxanthine

IH: antioxidant

LH: doubly allylic hydrogens from lipids

5

MBQ: methyl-l,4-benzoquinone

MeOH: methanol

MES: 2-(N-morpholino)ethanesulfonic acid

MF: monofonnazon of NBT

MOPS: 3-(N-morpholino)propanesulfonic acid

NADP: nicotinamide adenine dinucleotide phosphate

NADPH: reduced nicotinamide adenine dinucleotide phosphate

NBT: nitro blue tetrazolium

NHE: nonnal hydrogen electrode

O2-: superoxide anion

PC: phosphatidylcholine

PFC: plastic fonned carbon

y-Pyr: 3-hydroxy-2-methyl-4-pyrone

SI: soyasaponin I

SOD: superoxide dismutase

a-Toc: a-tocopherol

Trolox: 2-carboxy-2,5,7 ,8-tetramethyl-6-chromanol

XOD: xanthine oxidase

6

General Introduction

Saponins, i.e. sapogenin glycosides, are widely distributed in plants and a great

number of saponins have been reported (Mahato et aI., 1992). Although their broad

pharmacological activities have been extensively studied (Mahato et aI., 1992), their

phYSiological roles in plants have remained largely unknown. Recently, in our laboratory, a r­

pyronyl-triterpenoid saponin, named chromosaponin I (CSI, [1]; Fig. 1), has been isolated

from pea seedlings (Pisum sativum L. cv Alaska) (Tsurumi et aI., 1991, 1992). The same

compound was also isolated from soybean (Glycine max L.) (Kudou et aI., 1992). Alkaline

hydrolysis of CSI produces 3-hydroxy-2-methyl-4-pyrone (r-Pyr, [2]) and soyasaponin I (SI,

[3]). Although SI had been known to be one of the major saponins in leguminous plants,

almost all SI was extracted in the conjugate form with the y-pyronyl moiety in these plants,

suggesting that SI is an artificial product derived from CSI (Tsurumi et aI., 1991, 1992; Kudou

et aI., 1992).

The concentration of CSI in pea seedlings was 3.2 and 2.3 mM in the hook and the root

tip, respectively (Tsurumi et aI., 1992). CSI is found in all parts of pea seedlings, but at higher

concentrations in the hook and the root tip, both of which include the meristematic regions, than

in other non-growing tissues (e.g., 0.3 mM in the second internode). This suggests that CSI

is localized in the cytoplasm. When the seedlings were extracted with a buffer, CSI appeared to

be associated with high molecular weight compounds. The molecular weight of the associated

compounds was estimated to be about 50 kD and the y-pyronyl moiety was involved in the

association. However, the association is so weak that the compounds have not been isolated in

a purified form.

In the present study, the molecular properties of CSI were studied electrochemically

and biochemically, and CSI was found to have a definite reducing power, comparable

to a-tocopherol (a-Toe). Biological reductants are known to work as electron donors and

as radical scavengers. For example, ascorbate reduces cytochrome b561 and the protein­

bound Cu2+ in dopamine fJ-monooxygenase in chromaffin granules (Friedman and Kaufman,

7

~C02~O

OH

HO H

o

~H20H

HO 0 OH

H

o

HO~ Me H

OHOH

~C02~O

OH

HO H

o

~H20H

HO 0 OH

H

o

HO~ Me H

OHOH

30 29 o

o

1

2

OH

Fig. 1. Chemical structures of [1] CSI, 3-0-[a-L-rhamnopyranosyl-(1-72)-fJ-D­

galactopyranosyl(1-72)-.8-D-glucuronopyranosyI0-7 )]-22-0-[3' -hydroxy-2' -methyl-

5' ,6' -dihydro-4' -pyrone(6' -7 )]-3fJ,22fJ,24-trihydroxyolean-12-ene; [2] y-Pyr, 3-

hydroxy-2-methyl-4-pyrone; [3] SI, soyasaponin I.

8

1965; Kelley et aI., 1990). Ascorbate also inhibits lipid peroxidation through the regeneration

of a-Toc from the tocopheroxyl radical (McCay, 1985; Niki, 1987). The electron-donating

activity of CSI was evaluated in Chapter I and the radical-scavenging activity was investigated

in Chapter 2.

Biological reductants have been classified into two groups; hydrophobic (e.g.,

tocopherol and ubiquinol) and hydrophilic (e.g., ascorbate and NADPH). CSI is a new type

of biological reductant and the author proposes a third group, amphipathic, for CSI, as it is

composed of hydrophilic sugars and a hydrophobic aglycone with a y-pyronyl group. The

susceptibility of CSI to oxidation may be the reason why CSI had been not isolated in its native

form until it was recently isolated in our laboratory. Electron-donating activity in the cells is

very important in maintaining the growth and development of plants. The study of this new

biological reductant will offer a new approach in clarifying the roles of saponins in plants.

9

Chapter 1

Electrochemical and Spectrophotometrical Studies on CSI

I. Introduction

A great number of saponins have been isolated from plants, but their reducing activity

has not been reported. To evaluate the reducing power, it is useful to measure the redox

potential (EO). However, it is impossible to determine unequivocally the EO value for CSI

because it decomposes on oxidation. In the present study, the reducing power of CSI was

estimated by cyclic voltammetry and with redox-reactions in aqueous solutions, and it was

compared with other biological reductants. Kinetics of the redox-reactions and the pH

dependence of the reactions were also analyzed. The characteristics of CSI were described and

the possible roles of CSI in vivo were discussed.

II. Materials and Methods

II-I. Preparation of CSI, SI, and Cyt c

CSI was isolated from pea seedlings (Pisum sativum L. cv Alaska) as described

previously (Tsurumi et al., 1992), but with slight modifications. Seedlings grown in the dark at

25°C for a week were homogenized with three-times their volume of cold MeOH/aqueous 5

mM sodium ascorbate/HOAc (800 : 200 : 1, v/v), and the homogenate was centrifugated at

8,000 x g for 10 min. The supernatant was concentrated under reduced pressure and

chromatographed in a Sephadex LH-20 column, and eluted with MeOH/aqueous 5 mM sodium

ascorbate/HOAc (1600 : 400 : 1, v/v). The eluate containing CSI was repeatedly subjected to

reversed-phase HPLC (TSK gel ODS-120T, Tosoh) and eluted with MeOH/H20/HOAc (1600:

400 : 1, v/v). The eluate was further purified in the same column eluted with

MeOH/H20/HOAc (1440 : 560 : 1, v/v) to obtain CSI, this purified CSI being kept at -80°C in

a solution of MeOH/H20/HOAc (1600 : 400 : 1, v/v). SI was prepared by alkaline hydrolysis

of CSI as described previously (Tsurumi et aI., 1992).

Horse heart Cyt c was purchased from Wako Pure Chemical Ind., Ltd. and its

10

oxidized form was prepared by addition of a small amount of K3Fe(CN)6, followed by

treatment with a DEAE-cellulose column to remove the oxidant and K4Fe(CN)6 formed during

oxidation of Cyt c preparation (Yamanaka and Kamen, 1967).

11-2. Chemicals

y-Pyr, Trolox, and MBQ were purchased from Aldrich Chemical Co. 2,5-DMBQ

was from Tokyo Kasei Kogyo Co., Ltd. NADPH was from Oriental Yeast Co., Ltd.

Analytical grade reagents of sodium L-ascorbate, L-cysteine monohydrochloride monohydrate,

urate, and BQ were from Wako Pure Chemical Ind., Ltd. All other chemicals used were of

analytical grade. The aqueous solutions used were prepared using distilled water.

11-3. Electrochemical measurements with a GC electrode and a PFC electrode

The acquisition and analysis of voltammetric data were performed with a

microcomputer-controlled system (Osakai et aI., 1989). A three-electrode system was

employed with a glassy carbon (GC; Tokai Carbon, GC-30S; surface area = 0.071 cmZ) or a

plastic formed carbon (PFC; Mitsubishi Pencil, PFC41 No.5; surface area = 0.071 cmZ)

working electrode, a platinum counter electrode, and a Ag/AgCI (saturated KCI) reference

electrode. The GC electrode surface was polished with 0.06 11m alumina and 0.25 11m

diamond slurry successively. The PFC electrode surface was polished with a lapping film

abrasive (Marutho, No. 4000). Unless otherwise noted, for each record of a voltammogram,

the electrode surface was freshly polished with the diamond slurry or the lapping film, followed

by washing in an ultrasonic field in the series with distilled water and EtOH, and finally with

distilled water. Test solutions were degassed with prepurified Nz gas prior to the voltammetric

measurements. The electrolytic cell was water-jacketed to maintain the temperature at 25°C.

The controlled potential electrolysis of CSI was carried out with a Hokuto Denko,

Model HA-501 potentiostat equipped with a Model HF-202D coulometer. GC fibers (Tokai

Carbon, GC-20) were used as the working electrode.

11

11-4. Spectrophotometric measurements

For the reaction of CSI with BQ, the stock solution of CSI was concentrated under

reduced pressure and dissolved in 0.1 M buffers (pH 4.7, citrate-phosphate buffer; pH

5.7-7.7, Na-phosphate; pH 8.7, Na-borate). Since CSI degraded gradually in the pH 8.7

buffer, the CSI solution of pH 8.7 was prepared just before use. A solution of 10 mM BQ was

prepared with a help of sonication. The reaction mixture contained 0.1 mM CSI and 0.1 mM

BQ (or MBQ, 2,5-DMBQ). To start the reaction, 20,u1 of the 10 mM BQ was added to 2 ml

of the CSI solution, and the decrease in absorbance at 245 nm (BQ), 250 nm (MBQ), or 256

nm (2,5-DMBQ) was measured.

For the reaction of Trolox or NADPH with BQ, the reaction mixture contained 0.1 mM

Trolox (or NADPH) and 0.1 mM BQ. The reaction was started by adding 20 ,ul of 10 mM

Trolox or NADPH solution to 2 ml of the BQ solution (pH 7.7), and the decrease in absorbance

at 245 nm was measured. For the reaction of ascorbate with 2,5-DMBQ, the reaction mixture

contained 0.5 mM sodium ascorbate and 0.5 mM 2,5-DMBQ. To start the reaction, 40 ,ul of

25 mM sodium ascorbate in water was added to 2 ml of 0.5 mM 2,5-DMBQ, and the decrease

in absorbance at 330 nm was measured.

The reduction rate of BQ was determined from the decrease in absorbance at 245 nm;

the molar extinction coefficient £245 (= 19,500) of BQ is far greater than that of CSI (£245 =

540) and Trolox (£245 = 530); the coefficient of NADPH (£245 = 9,500) was almost the

same as that of NADP. The reduction rate of 2,5-DMBQ was determined from the decrease in

absorbance at 330 nm, since ascorbate has no absorbance at 330 nm. Plotting the reciprocals of

the concentration of quinones against time provided straight lines for limited reaction periods as

shown in Fig. 11, and the apparent second-order rate constants were determined from the slope.

For the reaction of CSI with Cyt c (oxidized form), a CSI solution of 1.9 ml in 50 mM

MOPS buffer (pH 6.7) was prepared as described above, and the reaction was started by adding

100 ,ul of 1.0 mM Cyt c in the CSI solution. The effect of BQ was measured by adding 20

,ul of 5 mM BQ in 50 mM MES buffer, pH 5.7, to the CSI solution together with the Cyt c

solution. The concentrations of CSI, Cyt c, and BQ in the reaction mixture were all 50,uM.

12

The reduction of Cyt c was determined by the increase in absorbance at 550 nm (Ll£550

(reduced - oxidized) = 19,100) (Nakamura and Kimura, 1971).

UV-visible spectra were recorded on a Hitachi Model U-3210 spectrophotometer, and

on a Jasco Ubest-30 spectrophotometer.

11-5. Other measurements (HPLC and 02 uptake)

Since the changes in CSI concentration during the reaction cannot be followed by the

absorbance change of the reaction mixture including BQ, CSI was determined by HPLC using a

ODS column (Tosoh, TSK gel ODS-120T; 7.8 x 300 mm). An aliquot ofreaction mixture was

subjected to the column, eluted with a mixture of MeOH/H20/HOAc (850 : 150: 1, v/v) and

measured at 295 nm.

Oxygen concentration was measured with a Clark electrode, Rank Brothers Ltd., UK.

III. Results and Discussion

III-I. Electrochemical behavior of CSI at the GC electrode

Figure 2(A) shows cyclic voltammograms obtained with a GC electrode in the presence

of (a) 0.02, (b) 0.05, and (c) 0.2 mM CSI in 0.1 M phosphate buffer (pH 7.7), In any case, a

well-developed anodic peak appeared around 0.45 V, though the peak potential was somewhat

affected by the CSI concentration. It should be noted that the anodic peak was not followed by

a cathodic (negative-current) peak on the reverse scan, indicating that CSI was decomposed

immediately after the electrochemical oxidation. At voltage scan rates (v) less than 5.0 V s-1,

such an irreversible wave was always observed (data not shown).

As shown in Fig. 2(B), r-Pyr also gave an irreversible anodic wave. The anodic peak

potential of CSI was more negative by > 0.15 V than that of r-Pyr, suggesting that the

r-pyronyl group was more susceptible to oxidation than its free form. In contrast to r-Pyr, SI

gave no redox wave (data not shown). These results clearly indicate that the y-pyronyl moiety

served as an electroactive site in CSI. The anodic peak current of y-Pyr was proportional to the

13

A

0.3 flA I

B

- 0.2 o 0.2 0.4 0.6 0.8

E (V vs. Ag / Agel)

Fig. 2. Cyclic voltammograms of (A) CSI and (B) y-Pyr recorded with a GC electrode in

0.1 M phosphate buffer (pH 7.7). (A) CSI concentrations are (a) 0.02 mM, (b) 0.05 mM,

and (c) 0.2 mM. (B) y-Pyr concentration: 1.0 mM. Broken lines are base currents. The

vertical lines indicate the current scales (note the difference in A and B). The voltage scan

rate: 0.1 V s-l.

14

square root of v (0.02-0.4 V s-l) [Fig. 3(B)], indicating that the oxidation process was

controlled by the diffusion of y-Pyr to the electrode surface (Bard and Faulkner, 1980).

However, this electrode process was somewhat different from that of CSI as described below.

The cyclic voltammogram of CSI has characteristics of an adsorption wave (Bond,

1980). The anodic peak current (I pa) was not very dependent on the CSI concentration. As

shown in Fig. 4, Ipa reached a constant only at 0.005 mM and then increased gradually at the

concentrations> 0.2 mM. In addition, Ipa was proportional to v (not to v1!2 for a diffusion­

controlled process) in the range of 0.02 to 0.4 V s-l. These dependences may be elucidated in

terms of the adsorption of CSI on the electrode surface. The adsorption process seems to be

very rapid, because the peak currents were reproduced even if voltammograms were recorded

only one minute after dipping the electrode in the test solution.

In order to examine the adsorption behavior of CSI on the electrode surface, the

following experiments were performed: Onto a GC electrode surface, CSI was adsorbed by

dipping the electrode for one minute in a phosphate buffer (pH 7.7) containing a lower

concentration (0.04 mM) or a higher concentration (1.0 mM) of CSI. The CSI-adsorbed

electrode prepared in this manner was then transferred into a deaerated buffer solution and

allowed to stand for one minute with or without stirring (by bubbling N2). In each case, a

voltammogram was recorded after the solution became calm. Curve (a) in Fig. 5(A) and in

5(B), being obtained without stirring, was almost identical to that recorded in a CSI-containing

solution [curves (a) and (c) in Fig. 2(A)]. As for the electrode prepared with a dilute CSI

solution, the voItammograms were hardly influenced by stirring prior to the voltammetric

measurements [curves (a) and (b) in Fig. 5(A)]. These results showed that CSI was strongly

{l.e. Irreversibly) adsorbed on the electrode surface at lower concentrations. On the other

hand, the voltammogram obtained at the electrode prepared with a relatively concentrated CSI

solution was highly affected by stirring [compare curves (a) and (b) in Fig. 5(B)]. Since the

area of the anodic peak became smaller in curve (b), a part of CSI, being adsorbed weakly,

would be detached from the electrode surface by stirring; note that the area of the anodic peak is

proportional to the amount of electricity required for the electrolysis of the redox species

15

A B

1.0 1.0

-~ -

0.5 0.5

0.0 L...-__ --L.. ___ .L......I 0.0 L...-_......L..-_---1... ___ L--I

o 100 200 o 5 10 15

v (mV/s) v1/2(mV/s)1/2

Fig. 3. Plot of the anodic peak current (Ipa) against the voltage scan rate (v ) for

cyclic voltammetry. (A) CSI concentration was 0.04 mM. (B) y-Pyr concentration

was 0.04 mM. Experimental conditions as in Fig. 2.

16

0.6

• 0.5

• • 0.4 P • ..-

1 -ctl 0.3 ...e-

0.2

0.1

0.0 _..._---..JI....-----L--.....L...--..L..---I....-...J

o 0.1 0.2 0.3 0.4 0.5

CSI (mM)

Fig. 4. Effect of the CSI concentration on the anodic peak current (/pa).

Experimental conditions as in Fig. 2.

17

A

0.3 flA I ---- -- - - - --

B

0.3 ,.,A I

-0.2 o 0.2 0.4 0.6

E (V vs. Ag I Agel)

Fig. 5. Cyclic voltammograms showing the adsorption behavior of CSI on the GC

electrode surface. Onto a GC electrode surface, CSI was adsorbed by dipping the electrode

for one min in a phosphate buffer (pH 7.7) containing CSI at 0.04 mM (A) or 1.0 mM (B).

The CSI-adsorbed electrode was then transferred into an electrolytic cell containing a

degassed phosphate buffer and allowed to stand for one min without (a) or with (b) stirring

(by bubbling N2) before voltammetric measurements. Other conditions as in Fig. 2.

18

adsorbed at the electrode (Bond, 1980). These results suggest that CSI is adsorbed at the

electrode in two states; in one state, CSI exists in a monomolecular layer on the electrode

surface, and in the other state, multimolecular layers are possibly formed.

111-2. Coulometric measurement of CSI

In controlled potential electrolysis, it was not possible to determine unequivocally the

number of electrons (n) required for the oxidation of a CSI molecule, because of the unceasing

current due to the successive oxidation of the oxidation product(s). Nevertheless, n for the

initial oxidation step is estimated to be lager than 2.

A kinetic analysis of the reduction of 1,4-benzoquinone (BQ) by CSI has clearly

demonstrated that two electrons participate in the oxidation of CSI (111-5).

111-3. Voltammetric comparison of the oxidation potentials of CSI and

biological antioxidants

With a view to estimating the reducing power of CSI, voltammetric measurements were

extended to some biological antioxidants, viz. ascorbate, urate, and cysteine. All of these

compounds gave irreversible waves similarly to r-Pyr, the anodic peak currents being

proportional to v1/2• The anodic peak potentials (Epa) of the antioxidants, CSI, and y-Pyr at

pH 7.7, which were measured at a definite scan rate, i.e. 0.1 V s-l, are shown, in Table I,. In

general, the values of Epa for an irreversible wave are determined not only by the eqUilibrium

redox potential but also by the electrode reaction rate, and consequently affected by voltage scan

rate (Bond, 1980). Accordingly, a strict comparison of the Epa values for different compounds

is not pertinent, but the results in Table I are informative to a certain extent. As can be seen in

the table, the "apparent" oxidation potential of CSI was ca. 0.3 V more negative than that of

cysteine and close to those of ascorbate and urate, suggesting that the reducing power of CSI

was stronger than that of cysteine and comparable to those of ascorbate and urate.

19

Table I. Anodic peak potentials, Epa' oleS! and other biological antioxidants.

Cyclic voltammograms were recorded with a GC electrode in 0.1 M phosphate buffer (pH 7.7)

at 25°C. The potential scan rate was 0.1 V s-l.

Chemicals Epa/ V

0.02 mM CSI 0.42

0.05 mM CSI 0.45

0.20mM CSI 0.51

y:--Pyr 0.67

ascorbate 0.35

urate 0.40

cysteine 0.79

111-4. Reductions of Quinones and Cyt c by CSI

To make sure of its reducing power, CSI was allowed to react with BQ, and the

reaction was monitored by spectrophotometry. CSI reduces BQ at pH 7.7, as shown in Fig. 6.

The decreases in absorbance at 245 nm and at 295 nm correspond to the reduction of BQ and

the oxidative degradation of CSI, respectively. The absorption peak at 295 nm shifted to

around 285 nm due to the formation of oxidation products of CSI. In Fig. 7, the reducing

power of CSI (B) against BQ is compared with those of some reductants including ascorbate

(A), Trolox (C), and urate (D). Trolox [4] is a water-soluble analogue for a-Toc [5] and its

reducing power is known to be similar to a-Toc (Barclay et aI., 1984). The reducing activity

becomes stronger in the order: urate < Trolox < CSI < ascorbate (ascorbate-·, H+/ascorbate

monoanion).

20

Q) () r:::::: ~ .0 ~

o CJ)

.0 <C

2.0 r---r------r------r------,

1.5

1.0

250 300 350 400

Wavelength (nm)

Fig. 6. Spectral change for the BQ-CSI system. The reaction cell contained 0.1 mM BQ

and 0.1 mM CSI in 0.1 M phosphate buffer (pH 7.7). A spectrum was recorded every 5

min. The absorbances at 245 nm and 295 nm decreased. Measurements were made at room

temperature (23°C).

21

- urate ~ 10 0 --Q) Trolox u c: ro .0 ~

0 (/)

.0 «S c: Q) 50 (/)

«S Q) ~

u Q)

"'C ED > ascorbate .";:

«S Q)

a: 0 0 5 10

Time (min)

Fig. 7. Reductions of BQ by ascorbate (. ), CSI ([] ), Trolox (0 ), and

urate (A). The reaction cell contained 0.1 mM BQ and 0.1 mM

reductants (ascorbate, CSI, or urate) in 0.1 M phosphate buffer (pH 7.7).

Absorbance changes were followed at 245 nm.

22

This order is in hannony with the formal redox potentials (EO') of urate, Trolox, and ascorbate,

i.e., 0.590, 0.480, and 0.282 V VS. NHE at pH 7.0, respectively (Buettner, 1993).

Accordingly, the EO' -value for CSI may be expected to lie between 0.3 and 0.4 V.

4 5

In addition, the reductions of some quinone derivatives, i.e., MBQ and 2,5-DMBQ,

and Cyt c by CSI were examined. CSI could reduce MBQ and Cyt c, but could not effectively

reduce 2,5-DMBQ (Fig. 8).

Figure 9 shows the time courses of the reduction of Cyt c by CSI at a physiological

pH (6.7) in the presence and absence of BQ. CSI reduced Cyt c very slowly, but the reduction

rate was greatly enhanced in the presence of BQ. This was also confirmed at other pH's in the

range of 5.7 to 7.7. This clearly shows that BQ can serve as a mediator in the reduction of Cyt

c, as shown in Scheme I.

CSIX BQ X 2 Cyt c (Fe2+)

oxidation products HQ 2 Cyt c (Fe 3~ (via semiquinone)

Scheme I. Mediator function of BQ in the reduction of Cyt c by CSI.

Since the formal redox potentials (EO') for BQ, MBQ, 2,5-DMBQ, and Cyt c,

respectively, are known to be 0.280 V (Fultz and Durst, 1982),0.24 V (see 111-8 in this chapter),

0.180 V (Fultz and Durst, 1982) and 0.255-0.260 V (Eddowes and Hill, 1979; Buettner, 1993)

vs. NHE (pH 7.0). Thus, CSI can reduce the compounds with EOl ~ about 0.2 V.

23

---~ o --Q) o c ro .0 .... o en .0 ro c Q) en ro ~ o Q) "0 Q)

> ~ Q)

a:

100

80

• 60 •

40

20

• • • • • • •• A

\ "

B

ra ••• OL....----_l--____ l--____ ~

o 100 200 300

Time (min)

Fig. 8. Reductions of quinones by CSI. The reaction cell contained 0.1 mM quinones

(A, BQ; B, MBQ; C, 2,5-DMBQ) and 0.1 mM CSI in 0.1 M phosphate buffer (pH 7.7).

Absorbance changes were followed at 245 nm (BQ), 250 nm (MBQ), and 256 nm (2,5-

DMBQ).

24

6 ~----------------------~

5

-~ 4 ::t -0

Q)

E 3 0

~

..c () 0 +-' >. 0 2 "'0 Q) () ::J CSI "'0 Q) 1

a:: SO cont

0 0 5 10

Time (min)

Fig. 9. Reduction of Cyt c by CSI and its acceleration by BQ.

The reaction cell contained 50 ,uM Cyt c and 50 ,uM CSI in

50 mM MOPS buffer (pH 6.7) with or without 50 ,uM BQ.

CSI+BQ (0 ), CSI (.), BQ (0), and control (.). Changes in

absorbance were followed at 550 nm.

25

111-5. Kinetic analysis of the reduction of BQ by CSI

To see the stoichiometry for the CSI-BQ system, [CSI] was measured with the help of

HPLC, and [BQ] was measured directly from the absorbance at 245 nm. As shown in Fig. 10,

the decrease of CSI parallels that of BQ, giving a stoichiometry of 1 : 1 for the ratio of CSI and

BQ. Accordingly, the reduction of BQ by CSI may be expressed as

CSI + BQ ~ CSlox + HQ (1)

where CSlox represents the oxidation products of CSI, soyasaponin I and y-pyrone derivatives

of unknown structures. The apparent rate constant (kapp) for the redox reaction follow the rate

law:

_ d [A] = lea [A][D] d t pp (2)

where A = BQ and D = CSI. If the initial concentration of the donor (CSI) is equal to that of the

acceptor (BQ), i.e., [D]o = [A]o, this equation is rewritten as

(3)

Thus the rate law is second order in BQ or in CSI. In this case, the following equation can be

derived.

(4)

According to this equation, the reciprocals of [BQ] were plotted against time, as shown in Fig.

11B. As expected, the plots provided straight lines. From their slopes the apparent rate

constants can be determined at some different pH's; kapp = 0.11, 0.78, 5.22, and 73.70 M-l

s-1 at pH = 5.7,6.7, 7.7, and 8.7, respectively. Thus the reduction of BQ by CSI is accelerated

26

-E -o E c

-o Q) (/)

ro Q) ~ C,) Q) o

o 10 20 30 40 50 60

Time (min)

Fig. 10. Time-course of the reduction of BQ by CSI. The reaction cell

contained 0.1 mM BQ and 0.1 mM CSI in 50 mM MOPS buffer (pH 7.7).

Decrease of [CSI] (0) was determined by HPLC, and that of [BQ] (. )

was determined by the absorbance measurement at 245 nm.

27

in alkaline solutions, while it is inhibited in acidic solutions. Below pH 4.7, no unequivocal

reaction was observed.

As shown above, CSI reacts with BQ with a stoichiometry of 1 : 1, and BQ is a two­

electron donor. It may be then concluded that two electrons should participate in the redox

reaction of CSI and BQ. Since the BQ is a two-electron oxidant, the oxidation of CSI could

occur favorably via two successive one-electron-transfer reactions (5), (6), (7), and (8).

(5)

(6)

+ kpl A + D -7 A + products (7)

kp2 A - + D+ ~ d- + products (8)

If D+ could escape from the sol vent cage of A -, D+ must react with 02 as described in

Chapter 2. However, 02 consumption was not observed in the reaction of CSI and BQ (data

not shown). Therefore, it is reasonable that D+ remains in the solvent cage of A- and reacts

with A-according to the equation (8). Hence the reaction is simplified as the equation (9).

kf A + D ~ d- + products (9)

According to Marcus theory (Marcus, 1968; Williams and Yandell, 1982; Kimura and

Kaneko, 1984; Marcus and Sutin, 1985), the variation of the excess free energy of activation for

a cross reaction (~G*) with the standard free-energy change (~GO) for a redox step such as that

in equation (5) can be expressed by use of equations (10)-(13).

28

N 1.0

(A) (8) 30

I

pH 4.7 ~

5.7

6.7 25~ I pH 8.7 ..-,.... ,

~ --..-

~ C') 20 ~ 60 b 0 --LO T"""

'¢ X C\I

<C ,.... ,

40 8.7 .--. 0 cc 15 ........

I ~ 20

I - - -~-.......--",-- -

10 I

0 0 10 20 30 0 10 20 30

Time (min) Time (min)

Fig. 11. (A) Decrease of A245 of BQ and (B) plotting of the reciprocals of [BQ] against time on the reduction of BQ by CSI at various pH's. The reaction cell contained 0.1 mM BQ and 0.1 mM CSI in 0.1 M buffer (pH 4.7, citrate-phosphate; pH 5.7-7.7, Na-phosphate; pH 8.7, Na-borate).

k = Z exp (-t1.G*/RT)

t1.G* = W + A (1 + t1.Go'/A)2/4

t1.GO' = t1.GO + (Op - (Or

A = 2 (t1.G * AA - (OAA + t1.G*DD - rooD)

(10)

(11)

(2)

(13)

Here Z is the collision frequency in solution (usually taken as 1011 dm3 mol-1 s-1 at 25°C), t1.

G* AA and t1. G*DD are the free-energy of activation for the self-exchange electron-transfer

reactions, WAA and rooD represent the work terms involved in the same reactions, ~ and Wr are

the work terms required to bring the reactants or products together in the activated complex,

and A = AO + Ai where AO is the solvent reorientation term and Ai represents the contribution

from the changes in inner-sphere bond lengths and angles in the activated state. By assuming

the adiabatic electron transfer and neglecting the work terms from (11)-(13), k (= kapp/[BQ],

1.58 dm3 mol-1 s-1, at pH 7.0) can be approximately given by (4)-07).

log k = 0/2)log K + 0/2)log kAAkDD (4)

= (-0.5 /2.3RT)t1.Go + (1/2)log kAAkDD (5)

= (0.5nF/2.3RT)M;0 + (1/2)log kAAkDD (6)

= 8.46(EO AA - EODD) + 0/2)log kAAkDD (7)

where K is equilibrium constant, kAA and kDD are the self-exchange rate constants for

BQ/BQ-. (chose to use a value of 106 dm3 mol- 1 s-1 [Williams and Yandell, 1982)] and

CSI/CSI+·, EO AA and EODD are the formal potentials for BQ!BQH. [-0.08 V VS, NHE, at pH

7.0 (Carlson and Miller, 1985)] and CSI/CSI+· (here, EODD = 0.3 V (IV-I), n is the number

of electrons involved in the reaction of CSI/CSI+· with BQ!BQ-· (here is n = 1), T is Kelvin's

temperature (here, T = 298K). The kDD value was estimated by using equation (17) to give

10-5 dm3mol- 1s-1• Since this value is too low as the self-exchange rate constants, the reaction

(5) may be not obeyed to Marcus cross relation in outer-sphere electron-transfer. This result

suggests that the reaction rate constants for the CSI-BQ system probably depends on

protonation and conformation change of CSL

30

111-6. pH Dependence of kapp for CSI and the comparison with those for other

reductants

Similar kinetic runs for the reduction of the quinones were carried out with other

reductants, viz. ascorbate, Trolox (a water-soluble analogue of a-Toc), and NADPH, in the

pH range of 4.7-8.7. The kapp values determined for the reductants are compared with those

for CSI in Fig. 12. As seen, all the log kapp vs. pH plots show straight lines with the

slopes of ~log kapp/~pH = 0.93, 0.35, 0.025, and -0.28 for CSI, ascorbate, NADPH, and

Trolox, respectively. For all the reductants except for ascorbate, BQ was used as the acceptor.

For ascorbate, the reduction rate of BQ was too fast to be measured, therefore 2,5-DMBQ was

employed as the acceptor in the place of BQ. The pH dependence of the redox potentials for BQ

and 2,5-DMBQ was almost the same, as will be shown below (Fig. 17). Hence, the pH

dependence of kapp shown in Fig. 12 should correspond to those of reducing powers of the

reductants. Thus, CSI shows the strongest pH dependence among the reductants tested for the

reduction of the quinones. It should be noted that the kapp for CSI at pH 8.7 is three orders of

magnitude greater than that at pH 5.7.

The mechanism in equations (18), (9), and (19) accounts for the results shown in Fig.

12, where DH = reductants, A = quinones. From these equations, the equations (20) and (21)

can be derived. Under the condition [H+] »Ka• log kapp linearly depends on pH.

K DH P D + H+ (18)

kf 2-A + D --7 J\ + products (9)

kf' A + DH --7 A2

- + products (19)

d[A] { Ka } - -d- = kr + (kf - kr) [A][D + DH] t Ka + [H+]

(20)

kapp = kr + (kf - kr) K a + Ka + [H ]

(21)

31

2.0r--------------------------. •

1.5

- 1.0 ..-'U>

..-, ~ -- 0.5

Q. Q. C1l

...:.:: 0>

0.0 0 ...J

Trolox

-0.5 CSI

-1 .0 L--__ L------::::...........JL.....-.._----L __ ----'-__ ---I

4 5 6 7 8 9 pH

Fig. 12. pH Dependence of the apparent second-order rate constants (kapp) for the

reductions of BQ by CSI (. ), NADPH (0 ), and Trolox ([] ), and for the reduction

of2,5-DMBQ by ascorbate (~). The reaction cell contained 0.1 mM BQ and 0.1 mM

reductants (CSI, NADPH, and Trolox), or 0.5 mM 2,5-DMBQ and 0.5 mM ascorbate

in 0.1 M buffer (pH 4.7, citrate-phosphate; pH 5.7-7.7, Na-phosphate; pH 8.7, Na­

borate). Data for ascorbate were presented only in the range of pH 4.7-6.7, because

at pH's 7.7 and 8.7 the reaction consumed oxygen and ascorbate was destroyed

rapidly without concomitant reduction of the quinone.

32

111-7. pH Dependence of Epa for CSI

Figure l3 shows the pH effects of cyclic voltammograms of CSI at a GC electrode.

With increasing pH, the anodic peak potential (Epa) due to the oxidation of CSI shifted to more

negative potentials with the rate of -90 mY/pH; Epa = 0.596,0.504, and 0.416 V at pH = 5.7,

6.7, and 7.7, respectively. This pH dependence is about three times stronger than that of the

formal potential of ascorbate, i.e., -30 mY/pH in the pH range of 6 to 8 (lyangi et aI., 1985).

This result is in harmony with the pH dependence of kapp (Fig. 12) for the reduction of

quinones by CSI and ascorbate.

III-S. Electrochemical behaviors of quinones at the PFC electrode

In order to evaluate the redox potentials of the electron acceptors (BQ, MBQ, and 2,5-

DMBQ) for CSI, cyclic voltammetric measurements were performed by using GC and PFC

electrodes. The electrochemistry of BQ has been poorly defined at ordinary electrodes such as

GC, pyrolytic graphite, gold, and platinum electrodes. In accord with the report by Kinoshita et

a1. (1994), the quinones gave ill-defined voltammetric waves at a GC electrode, but well­

defined, quasi-reversible voltammetric waves at a PFC electrode. The PFC electrode,

developed by Kawakubo et a1. (1992), produced quasi-reversible waves for the redox reactions

of MBQ and 2,5-DMBQ as well as BQ, indicating that the electrode is suitable for the

electrochemical detection of quinones. A typical example is shown in Fig. 14. The anodic and

cathodic peak heights were proportional to the square root of the voltage scan rate in the range

between 10 and 100 m V s-1 (Fig. 15), and also proportional to the concentration of BQ from

0.1 to 1.0 mM (data not shown). The midpoint potential (Emid) between the cathodic and

anodic peaks was 0.050 V vs. Ag/ AgCl at pH 7.7 (corresponding to 0.28 V vs. NHE at pH

7.0), which agrees well with the formal potential of 0.280 V vs. NHE at pH 7.0 (Fultz and

Durst, 1982). The peak separation of 42 m V is slightly larger than the expected value of 30 m V

for the reversible wave (n = 2). Similar voltammograms were obtained in the pH range

between 4.7 and 8.7, and also for MBQ and 2,5-DMBQ (data not shown). As shown in

Fig. 17, the plots of Emid's for BQ, MBQ, and 2,5-DMBQ against pH gave an identical slope,

33

1.0 pHS.?

I 0.5

0.0

-0.5 '---_--L.... __ .J.--_---'-__ ...J....-_--'-__ --'-_----'

-0.2 o 0.2 0.4 0.6 0.8

E (V vs. Ag I Agel)

Fig. 13. pH Dependence of cyclic voltammograms of CSI recorded with a GC electrode.

The electrolytic cell contained 0.04 mM CSI in 0.1 M phosphate buffer (pH 5.7-7.7). The

voltage scan rate was 0.1 V s-l.

34

30

20

10

r--.

< 0 :::t

........

""; -10

-20

-30

-40 -0.29 -0.20 -0. 10 0.00 0.10 0.20 0.29 0.40

E (V vs. Ag/AgCl)

Fig. 14. Cyclic voltammograms of BQ recorded with a PFC electrode (solid line) and a GC

electrode (broken line). The electrolytic cell contained 1.0 mM BQ in 0.1 M phosphate buffer

(pH 7.7). The voltage scan rate was 25 mV s-l.

35

40

20

,...., 0 <t: :::t '-"

h; -20

-40

-60

-0.20 -0.10 0.00 0.10 0.20 0.29

E (V vs. Ag/AgCl)

Fig. 15. Effect of voltage scan rates (v) on cyclic voltammograms of BQ recorded

with a PFC electrode. v = (A) 100; (B) 50; (C) 25; (D) 5 mV s-l. Other conditions as in

Fig. 14.

36

10

5

-5

-10

-15 ~------~------~------~------~------~--~ -0.20 -0.10 0.00 0.10 0.20 0.29

E (V vs. Ag/AgCl)

Fig. 16. pH Dependence of cyclic voltammograms of BQ with a PFC electrode. The

reaction cell contained 1.0 mM BQ in 0.1 M phosphate buffer (pH 5.7-7.7, Na-phosphate;

pH 8.7, Na-borate), The voltage scan rate was 5 mV s-l,

37

0.2 --.. 0 0> « -0>

0.1 « CJ)

> > ----"0

~ 0.0

-0.1

4 5 6 7 8 9

pH

Fig. 17. pH Dependence of the midpoint potentials (Emid) for

quinones with a PFC electrode. The electrolytic cell contained (A)

1.0 mM BQ, (B) 1.0 mM MBQ, or (C) 1.0 mM 2,5-DMBQ in 0.1

M buffer (pH 4.7, citrate-phosphate; pH 5.7-7.7, Na-phosphate;

pH 8.7, Na-borate).

38

which is close to the theoretically expected value of 59 mV (Pething et aI., 1983). The £0'­

values for MBQ and 2,5 DMBQ are 0.22 and 0.17 V vs. NHE, respectively at pH 7.0.

39

I. Introduction

Chapter 2

Antioxidative Effects of CSI

All living systems must contain self-protecting mechanisms. One such mechanism

concerns the protection against oxidative damage. Low molecular weight natural reductants

such as ascorbate and a-Toc have been shown to function as radical scavenging antioxidants

(McCay, 1985; Niki, 1987). Ascorbate protects nucleic acids and proteins from oxidative

damage (Fraga et aI., 1991), and it also inhibits lipid peroxidation through the regeneration of

a-Toc from the tocopheroxyl radical (McCay, 1985; Niki, 1987). Tocopherol (Burton et al.,

1983) and ubiquinol (Frei et al., 1990) are potent antioxidants in biomembranes. In humans,

urate, as well as ascorbate, is an important antioxidant in plasma (Ames et ai., 1981; Wayner

et aI., 1987) and nasal secretions (Peden et aI., 1990). In the previous chapter, CSI was

found to be a new biological reductant. In this chapter, then, the potency of CSI as an

antioxidant is described.

Some triterpenoid saponins have been shown to have pharmacological activities; for

example, protection of clinical ischemic and reperfusion injury (Nagai et al., 1992), inhibition

of inflammation (Ohuchi et ai., 1985; Amagaya et al., 1984) and lipid peroxidation (Ohminami

et al., 1984). Since these symptoms, as well as lipid peroxidation, involve free radical

production (Zweier et al., 1987; Bolli et al., 1989), the saponins' pharmacological effects may

be related to scavenging radicals. However, the direct evidence to show the ability of the

saponins to scavenge radicals has not been reported.

The ability of natural reductants to scavenge radicals has been investigated by using

stable free radicals, galvinoxyl (Takahashi et aI., 1986) and a,a-diphenyl-f3-picrylhydrazyl

(DPPH) (Blois, 1958), and a radical initiator, 2,2' -azobis(2-amidinopropane) dihydrochloride

(AAPH) (Niki et aI., 1986). The solution of DPPH shows a deep violet color. On accepting an

electron, the absorption vanishes and the resulting decolorization is stoichiometric with respect

to the number of electrons taken up (Blois et al., 1958). The water soluble AAPH generates

40

carbon-centered radicals at a constant rate by its thennal decomposition. They rapidly interact

with oxygen to give peroxyl radicals. The peroxyl radicals attack lipids and abstract doubly

allylic hydrogens from lipids to initiate free radical chain oxidation. The presence of

antioxidants reduces oxygen consumption and inhibits lipid peroxidation (Niki et al., 1986). In

this chapter, radical-scavenging activity of CSI was investigated by these methods.

II. Materials and Methods

11-1. Preparation of PC

Soybean phosphatidylcholine (PC) was obtained from Sigma Chemicals. A MeOH

solution of soybean PC was purified in alumina and silica gel columns before use (Kohen et al.,

1988; Niki et aI., 1986; Gotoh and Niki, 1992). The major polyunsaturated fatty acids in the

fatty acid composition of soybean PC are linoleic (68%) and linolenic (4%) acids (Goto and

Niki, 1992).

11-2. Chemicals

Galvinoxyl was purchased from Aldrich Chemical Co. AAPH, DPPH, and nitro blue

tetrazolium (NBT) were from Wako Pure Chemical Ind., Ltd. Hypoxanthine (HX), xanthine

oxidase (XOD), and superoxide dismutase (SOD) were purchased from Sigma Chemicals (St.

Louis, MO). 2-Methyl-6-phenyl-3,7-dihydroimidazo[I,2-a]pyrazin-3-one (CLA) was obtained

from Tokyo Kasei Kogyo. Other chemicals are described in Chapter 1.

11-3. Reaction with a stable free radical, galvinoxyl, in EtOn

A reaction cell contained a solution of 20 pM galvinoxyl [6] in EtOH. The reaction

was started by adding additives. Final concentrations of additives were 50 pM. Absorbance

change at 429 nm was measured.

41

0-

6

11-4. Reaction with a stable free radical, DPPH, in aqueous EtOH

A reaction cell contained 1 ml of 50 mM Na-MOPS buffer (pH 6.7 and 7.7) or Na­

MES buffer (pH 5.7), 1.25 ml of EtOH and 0.25 ml of 0.5 mM DPPH [7] in EtOH. Reaction

was started by adding reductants. Final concentrations of DPPH and reductants were 50 ,uM.

Absorbance change at 525 nm was measured.

<A. N-

CI 7

In the experiments to determine the stoichiometry between DPPH and reductants, the

final concentrations of DPPH and reductants were 100 ,uM and 10 ,uM, respectively. The

absorbance at 525 nm was measured 20 min after adding the reductants. In the experiments

under anaerobic condition, the volume of the reaction solution was 8 ml. The mixture was

bubbled with N2 for 10 min before adding reductants, and it was bubbled continuously after the

reaction started. The volume of the reaction solution was adjusted by adding EtOH before

measuring absorbance.

O2 consumption was measured with a Clark electrode (Rank Brothers Ltd., UK). The

reaction cell contained 2 ml of 50 mM Na-MOPS buffer (pH 7.7), 1 ml of EtOH and 1 mlof

42

0.5 mM DPPH in EtOH. The reaction was started by adding 100 /11 of CSI in MeOH. The

final concentrations of DPPH and CSI were 125,LIM and 100,LIM, respectively.

11-5. Reaction with a radical initiator, AAPH, in aqueous solution

A reaction cell contained 0.1 mM reductants, 10 mM AAPH and 0.1 M Na-phosphate

buffer (pH 7.4). The reaction was started by adding AAPH and the temperature of the cell was

kept at 3TC. The decrease of CSI, SI and Trolox was determined by HPLC and that of urate

and y-Pyr was measured spectrophotometrically. CSI, SI and Trolox were analyzed with an

ODS column (Tosoh, TSK gel ODS-120T; 7.8 x 300 mm) using mixtures of

MeOH/H20/HOAc (850 : 150: 1,800: 200: 1, and 500: 500 : 1, v/v) as eluants and measured

at 315,210, and 290 nm, respectively.

The azo compound AAPH generates free radicals at a constant and known rate as

described in 11-6 (Kohen et al., 1988).

11-6. Inhibition of the formation of PC-OOH in the AAPH system

The oxidation of PC (5.15 mM) as multilamellar liposomes in an aqueous dispersion

was also carried out in the presence of 2 mM AAPH at 3TC under air. The reaction solution

was prepared by adding 5 ml of MeOH to 1 ml of a solution of PC in MeOH (40 mg/m1), the

solvent then being removed by evaporating on a water aspirator with a rotary evaporator to

obtain a thin film of Pc. A 50 mM Tris-HCI buffer (10 ml, pH 7.4) was then added, and the

PC film was slowly peeled from the flask by shaking to obtain a white, milky liposome solution.

To start the reaction, 20/11 of a solution of AAPH in a Tris-HCl buffer was added to 2 ml of

the liposome solution. The oxidation of soybean PC can be measured by the formation of PC

hydroperoxide (PC-OOH). The reaction was carried out at 3TC with 2 mM AAPH, 5.15 mM

soybean PC liposomes, and 0.1 mM additives. The PC-OOH produced was analyzed by

injecting at various time intervals 10 /11 reaction mixture into an HPLC system, using a silica

gel column (Supelco, LC-Si; 4.6 x 250 mm) with a mixture of MeOH/40 mM phosphate (90 :

10, v/v) as an eluant at a flow rate of 1 ml/min, and detecting by UV-absorbance at 234 nm.

43

The oxidation of lipids initiated with AAPH (A-N=N-A) proceeds by the following

mechanism (Niki, et aI., 1986).

Initiation:

A-N=N-A ~ [A·N2·A] ~ (l-e) A-A

~ 2eA·

A02• + IH ~ stable products

A02• + A02• ~ stable products

A02• + LH ~ AOOH + L·

Propagation:

L· + 02 ~L02·

L02• + LH ~ LOOH + L·

Termination:

2L02• ~ non-radical products

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)

The water soluble AAPH generates geminate radicals by its thermal decomposition, a

portion of which recombine within the solvent cage (22) but rest of them escape from the

solvent cage and become free radicals (22). They rapidly interact with oxygen to give peroxyl

radicals (23). The peroxyl radicals are either scavenged by an antioxidant (lH) that is present in

the aqueous phase (24), interact with another peroxyl radical to give stable products (25), or

attack liposome lipids (LH) (26) to initiate free radical chain oxidation (27, 28).

11-7. Determination of O2- with NBT and CLA

(1) NBT method

A reaction cell contained 100 pM CSI, 10 mM AAPH and 100 pM NBT in 50 mM

Tris-HCI buffer (pH 7.8). The reaction was started by adding AAPH and the temperature ofthe

cell was kept at 3TC. The reduction of NBT was measured spectrophotometrically (530-560

nm).

44

Superoxide ( °2-) reduces NBT to give monofonnazon (MF) and difonnazon (DF) by a

sequence of the following 4-step reactions (30)-(33) (Gotoh and Niki, 1992). Monofonnazon is

blue and has the maximum absorption at 530 nm, while DF is blue-black colored and has the

maximum absorption at 560 nm.

NBT2+ + 02 - -7 NBT+ + 02

NBT+ + NBT+ -7 MF + NBT2+

MF+ + 02- -7 MF·

2MF· -7 MF+ + DF

(2) CLA method

(30)

(31)

(32)

(33)

To detect °2- produced in the CSI-AAPH system, the final concentrations of CSI,

AAPH, CLA and SOD were 50 ,lIM, 200 ,lIM, 0.17,uM and 300 unit/ml, respectively,

according to the method reported by Gotoh and Niki (1992). In the HX-XOD system the final

concentrations of HX, XOD, CSI and CLA were 100 ,lIM, 0.05 unit/ml, 25,uM (or 50,uM),

and 3.6 ,lIM, respectively. CLA, a Cypridina luciferin analogue, reacts with °2- to emit

chemiluminescence. At these concentrations, a mixture of HX, XOD and CLA produces a

constant chemiluminescence intensity for at least the first 10 min after the mixing, which is

essential for a kinetic analysis. All the reactions were carried out in a 50 mM Tris-HCI buffer

(pH 7.8 or 6.8) under air at 2Ye. The chemiluminescence intensity was measured with a single

photon counting apparatus, type OX-7, manufactured by Tohoku Electric Industries (Gotoh and

Niki,1992). This method has been extensively studied to detect 02- (Gotoh and Niki, 1992).

III. Results and Discussion

111·1. Reaction with a stable free radical, galvinoxyl, in EtOH

Galvinoxyl is a hydrophobic stable phenoxyl radical and is known to react with good

hydrogen donors such as tocopherols, ascorbic acid, cysteine, and glutathione (Takahashi et aI.,

1986). The reactivity of hydrogen donors toward galvinoxyl gives the relative activity

45

= = =>

_____ L _______ L-_____ L---:-----f--OO.

. , I :: !. : !

e c

,,' /' ~-"1 , ,. I , ,

: f. . i ~

! I "'~ .: i . . I ~ : . r .. ,-~.-:,. -;-:. -... -.---. ,i· 8

I . i : . j ~.' ...

·f . r i i. , :.

l! i 1/ ' . i .;, . f : .. - ~ -r--:-'~r:

j Ii i .; :,; I : .

J iii , i i : :

f. i ! : : .' i i' " : . / '; , J : :

I ~'~~.":.::I--~+ . : . ! . .,

. ,

;V ii~

III " 't It

, tl'l' I !! : { i (; , )'

J ' I I . , I

~,'. : I '. . , . 1J·· ., l' , : . I ~ 'i·'

-.J.._' r ,---. --'--r-= = = = = = <='J = LC> . .

CD CD =>

Absorbance

oj_

l/ .1

. ~ .. , .. !_- ,," l

CD

- - .,;. .. ~- ~--

CD CD CD

CD => CD

-CIJ -Q)

E r-

Fig. 18. Reactions of CSI, SI, y-Pyr, and a-Toe with a stable free radical, galvinoxyl.

The reaction cell contained 20 ,uM galvinoxyl, 50 ,uM additives (CSI, SI, y-Pyr, and

a-Toe), in EtOH. The reaction was started by adding additives. The reduction of

galvinoxyl was measured by the decrease of A429·

46

as a chain breaking antioxidant. Figure 18 shows that CSI interacts slowly with galvinoxyl,

whereas a-Toc reacts immediately. The figure also shows that y-Pyr hardly reacts and SI

dose not with galvinoxyl. These results indicate that the interaction of CSI with the radical is

very weak compared with a-Toc in EtOH solution.

111-2, Reaction with a stable free radical, DPPH, in aqueous EtOH

Since the radical-scavenging activity of natural reductants has been investigated in the

reaction with a stable free radical, a,a-diphenyl-,8-picrylhydrazyl (DPPH) (Blois, 1958), the

activity of CSI was measured by this method to compare with SI, y-Pyr and other reductants.

CSI reduced DPPH effectively but SI did not at pH 7.7 (Fig. 19). y-Pyr reacted with DPPH

but the rate was very low. These results clearly indicate that the y-pyronyl moiety is

responsible for scavenging the radicals. Figure 19 also shows that the radical-scavenging

activity of other biological reductants at pH 7.7. Ascorbate and Trolox were most active,

followed by cysteine, urate, CSI, glutathione and NADPH. The activity of CSI to donate

electrons depended on pH, stronger at pH 7.7 than in the lower pH ranges (Fig. 20).

The stoichiometric relation in the DPPH-CSI system was determined by calculating the

decrease of absorbance (~A 525) of 100 .uM DPPH on reacting with 10 .uM CSI, and

comparing them with that of ascorbate and cysteine (Table 11). According to the equations (34)

and (35), one mol of ascorbate and of cysteine are known to react with 2 mol and 1 mol of

ascorbate + 2 (DPPH)' ~ dehydroascorbate + 2 (DPPH):H (34)

2 cysteine(R-SH) + 2 (DPPH)' ~ R-S-S-R + 2 (DPPH):H (35)

DPPH, respectively (Blois, 1958). The stoichiometric determination revealed that one mol of

CSI reduced 1.1 mol of DPPH under aerobic condition but showed an increased reduction (1.7

mol of DPPH, closer to 2 as in ascorbate) under anaerobic condition. These results suggest that

CSI radicals reacted with O2 as well as DPPH under aerobic condition and that a part of CSI

radicals is involved in disproportionation. The reaction with O2 was supported by distinct O2

47

-~ 0 --LO C\I LO «

100

80

60

40

20

ascorbate (<» Trolox (+)

SI, r-Pyr

NADPH

glutathione

CSI

cysteine

O~----~----~~----~----~ o 20 40 60

Time (s)

Fig. 19. Reactions of CSI, SI, y-Pyr, and other reductants with a stable

free radical, DPPH. The reaction cell contained DPPH (50 ,uM) and the

indicated additive (SO ,uM) in a solution of EtOH/SO mM Na-MOPS buffer

(pH 7.7) (3 : 2, v/v). The reaction was started by adding an additive. The

reduction of DPPH was measured by the decrease of AS2S.

48

100~=-------------~

95

90

~ 85 o -80

75

70~~~~~~~-L~

a 4 8 12 16 20 Time (5)

Fig. 20. Effects of pH on the reduction of DPPH by CSI.

The reaction cell contained 50 pM DPPH in EtOH/50 mM

buffer (pH 7.7-6.7, Na-MOPS; pH 6.7, Na-MES) (3 : 2,

v/v) ..

49

CSI

40,uM 02 t

o 1 Time (min)

2

Fig. 21. O 2 uptake in the DPPH-CSI system. The reaction cell

contained 125 ,uM DPPH and 100,uM CSI in EtOH/50 mM Na-MOPS

buffer (pH 7.7) (1 : 1, v/v). O2 concentration was measured with a Clark

electrode and the reaction was started by adding CSI.

50

consumption in the DPPH-CSI system (Fig. 21). Ascorbate and cysteine did not show such

02 consumption in the reactions with DPPH (data not shown).

Table II. Stoichiometry in the reduction of 100 ,uM DPPH by lO,uM reductants (ascorbate,

CSI and cysteine) under aerobic and anaerobic conditions.

aerobic anaerobic

Reductant -------------------------------------------------------

ascorbate 0.238 ± 0.006

CSI 0.134 ± 0.007

cysteine 0.123 ± 0.003

(%)

100.0

56.3

51.6

n

2.0

1.1

1.0

0.225 ± 0.009

0.193 ± 0.005

0.114 ± 0.010

(%)

100.0

85.5

50.6

n

2.0

1.7

1.0

n is the number of electrons per reductant molecule involved in the reduction of DPPH.

The reaction period was 20 min. For the anaerobiosis the reaction solution was bubbled with

N2 continuously. Values are means ± s.e. for five experiments.

111-3. Reaction with a radical from AAPH in aqueous solution

In order to evaluate the radical-scavenging activity of CSI in aqueous solution, the

interaction with radical derived from a water-soluble radical initiator AAPH was investigated.

The rate of peroxyl-radical fonnation from the initiator is constant at a given temperature. In the

presence of AAPH, CSI, y-Pyr, Trolox and urate decreased linearly as shown in Fig. 22(A).

The decrease of CSI was most rapid and the rate was 3.25 ,uM min-I. The decrease rates of

both Trolox and urate were 0.30 ,uM min- I which corresponds to about a half of the radical

forming-rate from AAPH, i.e., 0.78 ,uM min- I estimated by Niki et al. (1986) under

51

U\ N

120rj ---------------------------, A 2.0 .a~,.....----------------__,

100 1.9 -0

- 80 1\ "e\. '),.. ~ 0>

::t 0

-- ~ 1.8 0> c

::t --C 0>

.(ij 60 c

E c Q)

.(ij

a: E 1.7 Q)

a: 40

I \ \.. 1.6

I \ '" 20 I 'i ~

1.5 0 50 100 150 200

0 1 ),.-0

0 50 100 150 200 Time (min)

Time (min)

Fig. 22. (A) Reactions of CSI and of other biological reductants with radicals induced by AAPH. The reaction cell contained 0.1 mM additives (or SI, r-Pyr), in 50 mM Tris-HCl buffer (pH 7.4) with or without 2 mM AAPH.

CSI+AAPH (0), r-Pyr+AAPH (~), SI+AAPH (A), Trolox+AAPH (D), urate+AAPH (-), CSI (.), SI (A.), and Trolox (-). The reaction was carried out at 37 ·C, and started by adding AAPH. (B) is plotting oflog[SI] against time.

the similar condition. The stoichiometric relation agrees well with the report describing that n

for both the molecules is 2 (Niki et aI., 1986). n is the number of electrons per reductant

molecule involved in the reduction of radicals. The decrease rate of r-Pyr, 0.66.uM min-I, is

two times greater than those of Trolox and urate, and it is almost equal to the rate of radical

formation, suggesting that n of r-Pyr is 1. Plotting of log [SI] against time provided a straight

line as shown in Fig. 22(B). The first-order reaction rate of SI with radicals was 0.53 x 10-2

min-I. However, SI did not show any antioxidative activity in the PC-AAPH system (111-4).

Hence, SI seems to react with the radicals as a substrate as observed in the reaction of PC with

radicals.

111-4. Antioxidative activity in the PC-AAPH system

Peroxidation of membrane lipids has been implicated as one of the primary events in

oxidative cellular damage. To evaluate antioxidative activity under biological conditions, it is

useful to measure the inhibitory effect on the oxidation of polyunsaturated fatty acids (Niki et al.,

1984). The rate of oxidation of PC in the absence of the antioxidant (control) was 15.3 nM

min- 1 (0% protection) (Fig. 23), while Trolox strongly inhibited the oxidation of PC (99.6%

protection). CSI showed antioxidative activity until 40 min (70.8% protection), the level of

inhibition being similar to that of urate (81.5% protection) and y-Pyr (70.8 % protection;

average value up to 60 min). But the antioxidative activity of CSI decreased with the elapse of

time and the oxidation rate of PC increased to 1.36.uM rnin- I beyond the control rate after 100

min (Fig. 23). The antioxidative activity of r-Pyr also decreased with the elapse of time and the

oxidation rate of PC approached the control level after 40 min, the average rate from 40 min to

180 min was 0.73 .uM min-I. SI exerted no antioxidative activity, the oxidation rate of PC

being 0.94.uM min-I.

53

180

160

140

120 --~ ::t. 100 ---I r-Pyr 0 0 80 I

() a..

60

40 urate

20 Trolox

0 50 100 150 200 Time (min)

Fig. 23. Antioxidative effects of CSI, SI, y-Pyr and other reductants (urate and

Trolox) on the PC-OOH formation in the oxidation of PC induced by AAPH.

Oxidation was carried out at 3TC, and the reaction mixture contained AAPH (2

mM) and soybean PC liposomes (5.15 mM) without an additive (control), or with

the indicated additive (0.1 mM). Control (+), SI ( • ), y-Pyr (. ), CSI (0 ), urate

(A) and Trolox (0 ).

54

111-5. °2- generation by CSI under aerobic condition

Mixing of CSI with NBT solution resulted in a gradual increase of the absorbance of

the mixture, and the solution became turbid [Fig. 24(A)]. After the increase of the absorbance

was almost saturated, a radical initiator AAPH was added into the mixture of NBT-CSI. As

shown in Fig. 24(B), absorbance at 530-560 nm of the mixture increased gradually after adding

AAPH, indicating that NBT was reduced to DF. The DF formation was inhibited by SOD

(Fig. 25). These results suggest that CSI radicals (CSI.) react with O2 to generate O2- (36)-

(38).

AAPH + 202 --7 2A02•

CSI- + A02• + H+ --7 CSI· + A02H

CSI· + 02 --7 stable products + °2-

CSI· + CSI· --7 CSI + stable products SOD

~- + 2H+ --7 ~~

(36)

(37)

(38)

(39)

(40)

Since CSI has a carboxyl group in the glucuronopyranosyl moiety, CSI molecules

dissociate under the neutral pH, and the anion form of CSI may aggregate with the NBT cations

according to the equations (41, 42), resulting in a turbid solution.

NBT2+ + 2CSI(Glut --7 complex (NBT aggregated with CSI) (41)

NBT+ + CSI(Glut --7 complex (NBT aggregated with CSI) (42)

Chemiluminescence intensity of the CLA solution increased immediately after adding

CSI [Fig. 26(A)], which was eliminated by SOD [Fig. 26(D), a dotted curve]. This result

suggests the presence of CSI radicals in the added-CSI solution and the formation of O2- due to

the reaction of the radicals with 02. When AAPH was added to the mixture of CSI-CLA

solution, the chemiluminescence intensity of the solution was immediately increased [Fig.

26(B)] and reached a steady level. The steady chemiluminescence was also decreased to almost

55

0.2

Q) u c ro .c 0.1 L-a C/)

.c <t:

o 480 540 620 660 480 540 620 660

Wavelength (nm)

Fig.24. Spectral changes (A) for the interaction of NBT with CSI and (B) for the reduction

of NBT in the presence of CSI and AAPH. The reaction cell contained 100 11M CSI, 100

pM NBT, and 10 mM AAPH in 50 mM Tris-HCl buffer (pH 7.8). A spectrum was recorded

every 1 min. Both the absorbances in (A) and (B) increased. The reaction was carried out at

3rC, and started by adding AAPH.

56

2.0 •

-- -SO~ E 1.5 c:: 0 CD L{) --CD

1.0 (.) c:: co .0

+ SOD .... 0 en .0 « 0.5

o 2 4 6 8 10

Time (min)

Fig. 25. Effect of SOD on the formation of diformazon from NBT in the

presence of CSI and AAPH. AAPH was added into the solution of NBT -CSI (. )

and NBT-CSI-SOD (0). SOD concentration is 300 units/ml. Experimental

conditions as in Fig. 24.

57

CLA + CSI (A)

>./ -·w c Q) -c Q) (,,) c Q) (,,) (IJ Q) c E ::J

E Q) .c ()

SOD (D)

. AAPH (8) .

CLA \ \ / ---.. ::". .... - ... ~ ... ~~~~~ ..................... . ~~~ background

o 1 2 3

Time (min)

Fig. 26. Increase of chemiluminescence from CLA by adding CSI and AAPH, and the

decrease of it by adding SOD. The final concentrations are 0.17 JiM CLA, 50 JiM CSI,

200 JiM AAPH and 476 units/ml SOD in 50 mM Tris-HCl buffer (pH 7.7). At (A) CSI

added to the CLA solution, at (B) AAPH added, and at (C) SOD added successively, and at

(D) SOD added to the fresh CLA+CSI solution.

58

-Q) > • .0= ctI Q) ~ --en c: Q) ..... c: Q) () c: Q) () en Q) c: E :::J

E Q)

..c: o

o 1

CSI

50,uM

25

o

background (-XOD, -HX)

234

Time (min)

5

Fig. 27. Effect of CSI on the chemiluminescence from CLA

induced by HX-XOD system at 25°C. XOD (0.05 unit/ml) was

added into the solution of 25 ,uM CSI (or 50 ,uM), 100 ,uM HX

and 3.6,uM CLA in 50 mM Tris-HCl buffer (pH 7.7).

59

zero by adding SOD, and AAPH alone did not increase the chemiluminescence of CLA solution

in the absence of CSI (data not shown). These facts indicate that CSI radicals formed in the

presence of AAPH react with 02 to produce O2- [Fig. 26(C)].

A mixture of HX -XOD-CLA solution emits chemiluminescence at a constant rate due to

02- production according to the equations (43)-(45) (Gotoh and Niki, 1992). The addition of

CSI into the mixture increased the chemiluminescence intensity of the solution (Fig. 27). Thus,

CSI did not show SOD activity, at least under aerobic conditions, in the aqueous solution at pH

7.7.

XOD HX + 02 ~ urate + 0i

°2- + CLA ~ products + chemiluminescence

02 - + X ~ products

60

(43)

(44)

(45)

Chapter 3

Implications and Perspectives

I. Characteristics of CSI as an Electron Donor

Since the redox reactions with CSI are irreversible (as other natural reductants

including ascorbate, cysteine, and urate etc.), the Eo' of CSI can be determined neither from

the reactions with quinones nor from voltammetric measurements. A tentative application of

Marcus theory (Marcus, 1968; Marcus and Sutin, 1985) to the CSI-<Iuinone reaction did not

enable us to determine the EO' value for CSI, probably owing to participations of the

protonation and conformation change in CSI. However, CSI can reduce BQ (EO' = 0.28 V

VS. NHE at pH 7.0) and MBQ (EO' = 0.22 V), but cannot effectively reduce 2,5 DMBQ

(EO' = 0.18 V) (Chapter 1). Also, CSI can reduce Cyt c [EO' = 0.255-0.260 V (Buettner,

1993; Eddowes and Hill, 1979)] effectively in the presence of BQ, as shown in Fig. 6.

Thus, it can be concluded that CSI has a reducing power which can reduce the compounds

with EO' 2:: about 0.2 V.

Since CSI is surface-active, as observed in the adsorption behavior at the GC

electrode surface, it is reasonable to expect that CSI shows certain affinity to biomembranes

(or to the hydrophobic moiety of proteins), and mediates an electron transfer between the

cytoplasm and the membrane (or the proteins). Biological reductants have been classified

into two groups: hydrophobic (e.g., tocopherol, ubiquinol and bilirubin) and hydrophilic

(e.g., ascorbate and urate). CSI should not be classified into either the hydrophobic or

hydrophilic group, but into a new "amphipathic" group, and the amphipathic reductant,

CSI, is expected to have a role different from other biological reductants in plant cells.

61

II. A Possibility of Participation of CSI as a pH Sensor in the Electron

Transfer Induced by pH Change in vivo

The pH dependence of the reducing power, especially its acceleration at alkaline pH's

(Fig. 12) is a very conspicuous characteristic of CSI. In my opinion, this strong pH

dependence is due to the dissociation of the hydroxyl group of the y-pyronyl moiety of CSI.

The dissociation constant could not be determined, because CSI is unstable at the pH's higher

than 8.7. However, since the pK value for 3-hydroxy-2-methyl-4-pyrone has shown to be ca.

8.5 by spectrophotometric measurements, the pK value for the y-pyronyl moiety of CSI may

be supposed to lie near the pH. This suggests that the strong pH-dependent reducing power

observed in the pH range of 4.7-8.7 is related to the dissociation of the y-pyronyl moiety. It

is reasonably assumed that the dissociated form of y-pyronyl moiety has stronger reducing

power than the undissociated form.

The change in the cytosolic pH has been proposed to be involved in hormonal signal

transduction in plants (Kurkdjian and Guern, 1989; Gehring et aI., 1990; von der Veen et aI.,

1992). Both auxin and abscisic acid were reported to increase cytosolic free-Ca2+ rapidly in

dark-grown com coleoptiles, parsley hypocotyls and their roots. The cytosolic pH was lowered

by auxin, while it was raised by abcisic acid (Gehring et aI., 1990). The cytosolic pH is

thought to influence the signaling by cytosolic free-Ca2+, possibly by regulating Ca2+-protein

binding. The contrasting effects of auxin and abscisic acid on cytosolic pH may underlie their

mutual antagonism (Gehring et aI., 1990).

CSI was found in the hook and the root tip of pea seedlings, both of which include the

meristematic regions, at higher concentrations, i.e., 3.2 and 2.3 mM per g fresh weight base,

respectively (Tsurumi et aI., 1992). In contrast, CSI was found in other parts at the lower

concentrations (e.g., 0.3 mM in the second internode, a non-growing tissue). This fact

suggests that CSI is localised in the cytoplasm. The cytosolic pH in plants has been reported to

lie between 6.8 and 7.5 (Kurkdjian and Guern, 1989). In this pH range CSI can reduce Cyt c,

and the reducing activity of CSI is comparable to that of the other known important reductants,

NADPH and Trolox (Fig. 12). NADPH is a coenzyme of oxidoreductases, and the tocopherol

62

has been known as a chain-breaking antioxidant for membrane lipids (Niki, 1987). These facts,

taken together, suggest the potential of CSI to function as an electron donor in vivo. As

described above, in the physiological pH, the reducing power of CSI changed greatly compared

with other reductants. This characteristic enhances a possibility of CSI as a pH sensor which

transfers the signal of pH-change in the cytoplasm to an acceptor in the membrane.

63

III. Antioxidative Effects of CSI and a Proposed Cyclic Mechanism

The antioxidative ascorbate has been argued to eliminate cytosolic free radicals via

hydrogen atom transfer (Scheme II, Njus and Kelley, 1991, 1993). According to this scheme,

at physiological pH, ascorbate (AH-) and semidehydroascorbate (A--) are present predominantly

in their anionic forms. This means that the transition between them involves the net gain or loss

of a hydrogen atom. Ascorbate monoanion is a poor electron donor because the monoanion has

a very high redox potential (E01 = +0.776 V) and the dianion (A2-) has an insignificant

concentration. The most common biological free radicals (OH-, ROO-, R-, etc.) are missing an

H atom rather than an electron. Consequently, they are acceptors of single H atom and are

reduced back to their parent compounds (H20, ROOH, RH) most effectively by a donor of

single H atom. The ability to donate single H atom accounts for the effectiveness of ascorbate

as a free radical scavenger (Njus and Kelley, 1993). CSI reacted with stable free radicals but

the reaction rates were conspicuously low compared with cysteine as well as with ascorbate and

a-Toe (111-1 and 111-2 in Chapter 2), although the Epa value for CSI was comparable to that of

ascorbate and more negative than that of cysteine (Chapter 1). This difference suggests that CSI

is an electron donor but not a single H atom donor. The reaction of CSI with stable free radicals

could be separated by two steps, i.e., the electron transfer from CSI to the radicals, and the

proton equilibration with the aqueous solvent. The reaction rate of CSI with radicals increased

with pH (111-2 in Chapter 2), indicating that the anionic form of CSI (CSI-) is the real reactant.

Since the y-pyronyl moiety is responsible for the reducing activity of CSI (Chapter 1 and 111-2

in Chapter 2), the dissociated form of the y-pyronyl moiety must be involved in the redox

reaction. A carboxyl group in the glucuronopyranosyl moiety of CSI dissociates also in

physiological pH, but this dissociation is not involved in the redox reaction and it was omitted

in Scheme II to avoid confusion. In Scheme II, CSI- means a CSI molecule dissociated in the

fLpyronyl moiety.

64

(A) CSI

CSI :::;e .... =========.~ csr

+e +e -e-E,02 > 0.3 V

CSI·

pKr 11 products

(8) Ascorbate pK1 = 4.0 pK2 = 11.3

-H + -H + AH2 :::;e'll(=========.~ AH-~'lI(=========·~ A2-

+e E01 = 0.776 V

-H + AH • :;;;jj'll(f==========~. A - •

EO" = ca. 0.3 V

E02= 0.076 V

pKr = -0.45 e-11 e-+ - E03= -0.210 V

A

Scheme II. Proton/electron transfer diagrams for (A) CSI and (B) ascorbate. Protonation reactions are shown horizontally and electron transfer reactions are arranged vertically. Abbreviations are: CSI-, CSI molecule dissociated in the I'pyronyl moiety; CSI+·, CSI radical cation; CSI·, CSI neutral radical; AH2, ascorbic acid; AH-, ascorbate monoanion; A2-, ascorbate dianion; A-·, semidehydroascorbate radical anion; AH·, semidehydroascorbate neutral anion; A, dehydroascorbate. Values for pK and EO in (B) are taken from Njus and Kelly (1993).

65

0.06 V

The dianionic form of ascorbate (A2-) is a strong electron donor and reduces molecular

oxygen to superoxide (Scheme III). However, alkalinization of CSI solution did not induce 02

consumption (data not shown), indicating that CSI- does not react with 02. Njus and Kelley

(1991) calculated that the redox potential of the 0i02-- couple in living cells should lie in the

range between +0.24 and +0.36 V. Hence, the £02 value for the CSI-/CSI- couple may be

more positive than +0.3 V. On the other hand, CSI radicals reacted with 02 under aerobic

condition. The generation of O2- was confirmed in AAPH system by both the methods with

NBT and CLA (111-5 in Chapter 2). From these results £03 for the oxidation of CSI- is

estimated tentatively to be the similar level with £02 (= 0.076 V) for the A2-/A -- couple. CSI

scavenged water-soluble radicals initiated with AAPH and hydrophobic stable free radicals, but

CSI was rapidly exhausted under aerobic condition. This consumption of CSI is probably due

to its autocatalytic oxidation with generating 02-. Although CSI was proved to have a potency

as an antioxidant in the PC-AAPH system, the antioxidative activity of CSI did not last longer

than one hour under aerobic condition probably due to the exhaustion of CSI. In this

experiment, however, the concentration of O2 in the reaction mixture was 230,uM (equilibrium

with the atmosphere) and the concentration of CSI was 100,uM. On the other hand, in living

cells the 02 concentration lies between 3 to 30,uM (Jones and Mason, 1978; Wittenberg and

Wittenberg, 1985) and the concentration of CSI is 2-3 mM (in the hook and the root tips).

Under these biological conditions in plants the generation of 02- is expected to be suppressed

profoundly and CSI could be a effective radical scavenger. CSI may also scavenge cation­

radicals more favorably, because only the transfer of electrons is necessary to scavenge these

radicals. Ascorbate is known to reduce tocopheroxyl radicals to tocopherol. Under the similar

mechanism CSI- could be reduced back to CSI by receiving single H atoms. In fact, CSI

oxidation by one electron oxidising enzyme, such as peroxidase, was perfectly suppressed in

the presence of 10 mM ascorbate (data not shown). In this case CSI molecules cycle along the

triangle as shown in Scheme III, i.e., CSI ~ CSI- ~ CSI- ~ CSI. This cycle makes

possible that CSI scavenges radicals with consuming ascorbate. The author proposes that this

cycle is essential for CSI to function as a radical scavenger in living cells.

66

(A) CSI

CSI-.......,.....-~»- csr "\

H+ a free radical

non-radical products

CSI· r---··.···--------·----··.---------··.-, r--------------_·-------------·_----------------_·_--- -----.--~

aerobic condition

0-o 2 :

a free radical

non-radical products

anaerobic condition

~ .. -------------.----------------------~ ~ _____________________________________________________________ J

decomposition

(8) Ascorbate

AH-"\

~ A2-

O2 H+

a free radical 0-

2

non-radical products

A-•

Scheme III. (A) Proposed cyclic mechanism (CSI ~ CSI- ~ CSI· ~ CSI) for the radical­

scavenging reaction by CSI in living cells. In the absence of ascorbate (dotted squares), CSI·

reacts with O2 under aerobic condition and with radicals under anaerobic condition. (B)

Alternative mechanisms of ascorbate oxidation. A2- reduces O2 to O2- by electron transfer.

AH- reduces radicals by H atom transfer. Abbreviations are as in Scheme II.

67

Some triterpenoid saponins have been shown to protect clinical ischemic and

reperfusion injury (Nagai et aI., 1992), and inhibit inflammation (Ohuchi et al., 1985; Amagaya

et aI., 1984) and lipid peroxidation (Ohminami et aI., 1984). Since the ischemic and

reperfusion injury and inflammation as well as lipid peroxidation involve free radical

productions (Zweier et al., 1987; Bolli et al., 1989), the saponins' pharmacological effects may

have some relation with scavenging the radicals. However, the direct evidence to show the

radical-scavenging activities of the saponins has not been reported. The present work clearly

indicated that CSI shows antioxidative activity to prevent peroxidation of PC as well as that

CSI scavenges stable free radicals. Ohminami et al. (1984) reported that SI inhibited the

aerobic peroxidation of corn oil. In the present work, however, SI did not prevent the

formation of PC-OOH initiated with AAPH. These results suggest that SI at higher

concentrations competes with corn lipids as substrates against radical attack. Hence, SI is not

an antioxidant and the excess SI is required to prevent lipid peroxidation. Okubo (1993)

reported that CSI showed a SOD-like activity by using NH30H method. In the present work,

however, the reaction of CSI with O2- was not proved. The NH30H method used by Okubo

to detect O2- is unspecific. Both the methods with NBT and CLA in the present work have

been shown to be more reliable to measure O2-.

The hydrophobic reductants (e.g., tocopherol and ubiquinol) are localized in

biomembranes to inhibit lipid peroxidation, although they are usually poor antioxidants for

water-soluble radicals. On the other hand, the hydrophilic reductants (e.g., ascorbate and

urate) are effective for water-soluble radicals, but are not efficient inhibitors of lipid

peroxidation by themselves because of their insolubility in lipids (McCay, 1985; Niki, 1987) .

Since CSI shows a hydrophobic property as observed in its adsorption behavior at the GC

electrode, it is reasonable to expect that CSI shows some affinity to biomembranes. The

hydrophobic contact of the membrane and the triterpenoid moiety possibly facilitate the electron

transfer from the r-pyronyl site to the peroxyl radicals (or cation radicals) in lipids, because

the peroxyl radical center of lipids is believed to float towards the surface of the bilayers due to

a large dipole moment (Barclay and Ingold, 1981; Fessenden et aI., 1984). Thus CSI is

68

expected to donate electrons to the radical species in lipids and to be reduced back to CSI by

receiving single H atom from the cytosolic ascorbate according to the cyclic hypothesis

proposed in this chapter.

69

References

Amagaya, S., Sugishita, E., Ogihara, Y., Ogawa, S., Okada, K., and Aizawa,

T. (1984) Comparative studies of the steroisomers of glycyrrhetinic acid on anti­

inflammatory activities. J. Pharm. Dyn. 7, 923-928.

Ames, B. N., Cathcart, R., Schwiers, E., and Hochstein, P. (1981) Uric acid

provides an antioxidant defense in humans againat oxidant- and radical-caused aging and

cancer: A hypothesis. Proc. Nat!. Acad. Sci. USA 78, 6858-6862.

Barclay, L. R. C., and Ingold, K. U. (1981) Autoxidation of biological molecules. 2.

The autoxidation of a model membrane. A comparison of the autoxidation of egg lecithin

phosphatidylcholine in water and in chlorobenzene.

J. Am. Chem. Soc. 103, 6478-6485.

Barclay, L. R. C., Locke, S. J., MacNeil, J. M., Van Kessel, J., Burton, G.

W., and Ingold, K. U. (1984) Autoxidation of micelles and model membranes.

Quantitative kinetic measurements can be made by using either water -soluble or lipid­

soluble initiators with water-soluble or lipid-soluble chain-breaking antioxidants.

J. Am. Chem. Soc. 106, 2479-2481.

Bard, A. J. and Faulkner, L. R. (1980) in Electrochemical Methods, Chapter 6, Wiley,

New York.

Blois, M. S. (1958) Antioxidant determinations by the use of a stable free radical.

Nature 181, 1199-1200.

Bolli, R., Jeroudi, M. 0., Patel, B. S., DuBose, C. M., Lai, E. K., Roberts,

R., and McCay, P. B. (1989) Direct evidence that oxygen-derived free radicals

contribute to postischemic myocardial dysfunction on the intact dog.

Proc. Natl. Acad. Sci. USA 86, 4695-4699.

Bond, A. M. (1980) in Modern Polarographic Methods in Analytical Chemistry, pp. 192-

196, Marcel Dekker, New York.

70

Buettner, G. R. (1993) The pecking order of free radicals and antioxidants: lipid

peroxidation, a-tocopherol, and ascorbate. Arch. Biochern. Biophys. 300, 535-543.

Burton, G. W., Joyce, A., and Ingold, K. U. (1983) Is vitamin E the only lipid­

soluble, chain-breaking antioxidant in human blood plasma and erythrocyte membranes?

Arch. Biochern. Biophys. 221, 281-290.

Carlson, B. W. and Miller L. L. (1985) Mechanism of the oxidation of NADH by

quinones. Energetics of one-electron and hydride routes.

J. Am. Chern. Soc. 107, 479-485.

Eddowes, M. J. and Hill, H. A. O. (1979) Electrochemistry of horse heart cytochrome c.

J. Am. Chern. Soc. 101, 4461-4464.

Fessenden, R. W., Hitachi, A., and Nagarajan, V. (1984) Measurement of the

dipole moment of a peroxyl radical by microwave dielectric absorption.

J. Phys. Chern. 88, 107-110.

Fraga, C. G., Motchnik, P. A., Shigenaga, M. K., Helbock, H. J., Jacob, R.

A., and Ames, B. N. (1991) Ascorbic acid protects endogenous oxidative DNA

damage in human sperm. Proc. Nat!. Acad. Sci. USA 88, 11003-11006.

Frei, B., Kim, M. C., and Ames, B. N. (1990) Ubiquinol-1O is an effective lipid­

soluble antioxidant at physiological concentration.

Proc. Natl. Acad. Sci. USA 87, 4879-4883.

Friedman, S. and Kaufman, S. (1965) 3,4-Dihydroxyphenylethylamine fj-hydroxylase.

J. Bioi. Chern. 240, 4763-4773.

Fultz, M. L. and Durst, R. A. (1982) Mediator compounds for the electrochemical study

of biological redox systems: a compilation. Anal. Chirn. Acta 140,1-18.

Gehring, C. A., Irving, H. R., and Parish, R. W. (1990) Effects of auxin and

abscisic acid on cytosolic calcium and pH in plant cells.

Proc. Natl. Acad. Sci. USA 87, 9645-9649.

71

Gotoh, N. and Niki, E. (1992) Rates of interactions of superoxide with vitamin E,

vitamin C and related compounds as measured by chemiluminescence.

Biochirn. Biophys. Acta 1115, 201-207.

Iyanagi, T., Yamazaki, I., and Anan, K. F. (1985) One-electron oxidation-reduction

properties of ascorbic acid. Biochirn. Biophys. Acta 806, 255-261.

Jones, D. P. and Mason, H. S. (1978) Gradients of O2 concentration in hepatocytes.

1. Bioi. Chern. 253, 4874-4880.

Kawakubo, T., Suda, Y., Kaneko, H., Negishi, A., and Yamada, M. (1992)

Electrochemical behavior of carbon microelectrodes prepared by using graphite/carbon

composite. TANSa 152, 106-114.

Kelley, P. M., Jalukar, V., and Njus, D. (1990) Rate of electron transfer between

cytochrome b561 and extravesicular ascorbic acid. 1. Bioi. Chern. 265, 19409-19413.

Kimura, M. and Kaneko, Y. (1984) Kinetics of electron-transfer reactions of para-

substituted phenols p-C6H4(X)OH with [Fe(phen)3p+(phen = 1,1O-phenanthroline) and

with [IrCI6]2- in aqueous acidic solutions: correlation between the Hammett constant of X

and the one-electron redox potential of p-C6H4(X)OH.

1. Chern. Soc. Dalton. Trans. 1984, 341-340.

Kinoshita, H., Suda, Y., Kawakubo, T., Takayama, K., and Ikeda, T. (1994)

Electroanalysis of NADH using a new electrode prepared from plastic formed carbon

(PFC). Microchern.l. 49,226-234.

Kohen, R., Yamamoto, Y., Cundy, K. C., and Ames, B. N. (1988) Antioxidant

activity of carnosine, homocarnosine, and anserine present in muscle and brain.

Proc. Natl. Acad. Sci. USA 85, 3175-3179.

Kudou, S., Tonomura, M., Tsukamoto, C., Shimoyamada, M., Uchida, T.,

and Okubo, K. (1992) Isolation and structural elucidation of the major genuine

soybean saponin. Biosci. Biotech. Biochern. 56, 142-143.

Kurkdjian, A. and Guern, J. (1989) Intracellular pH: measurement and importance in cell

activity. Annu. Rev. Plant Physiol. Plant Mol. Bioi. 40, 271-303.

72

Mahato, S. B., Nandy, A. K., and Roy, G. (1992) Triterpenoids.

Phytochemistry 31, 2199-2249.

Marcus, R. A. (1968) Theoretical relations among rate constants, barriers, and bnjmsted

slopes of chemical reactions. J. Phys. Chem. 72, 891-899.

Marcus, R. A. and Sutin, N. (1985) Electron transfer in chemistry and biology.

Biochim. Biophys. Acta 811, 265-322.

McCay, P. B. (1985) Vitamin: interactions with free radicals and ascorbate.

Ann. Rev. Nutr. 5, 323-340.

Nagai, T., Egashira, T., Kudo, Y., Yamanaka, Y., and Shimada, T. (1992)

Attenuation of dysfunction in the ischemia-reperfused liver by glycyrrhizin.

Japan. 1. Pharmacol. 58, 209-218.

Nakamura, S. and Kimura, T. (1971) Studies on spinach ferredoxin-nicotinamide adenine

dinucleotide phosphate reductase. Kinetic studies on the interactions of the reductase and

ferredoxin and a possible regulation of enzyme activities by ionic strength.

1. Bioi. Chem. 246, 6235-6241.

Niki, E. (1987) Antioxidants in relation to lipid peroxidation.

Chem. Phys. Lipids 44, 227-253.

Niki, E., Saito, T., Kawakami, A., and Kamiya, Y. (1984) Inhibition of oxidation

of methyllinoleate in solution by vitamin E and vitamin C.

J. Bioi. Chem. 260,4177-4182.

Niki, E., Sato, M., Yoshikawa, Y., Yamamoto, Y., and Kamiya, Y. (1986)

Oxidation of lipids. XII. Inhibition of oxidation soybean phosphatidylcholine and

methyl1inoleate in aqueous dispersions by uric acid.

Bull. Chem. Soc. Jpn. 59, 471-477.

Njus, D. and Kelley, P. M. (1991) Vitamins C and E donate single hydrogen atoms

in vivo. FEBS Lett. 284, 147-151.

Njus, D. and Kelley, P. M. (1993) The secretory-vesicle ascorbate-regenerating system: a

chain of concerted H+je--transfer reactions. Biochim. Biophys. Acta 1144,235-248.

73

Ohminami, H., Kimura, Y., Okuda, H., Arichi, S., Yoshikawa, M., and

Kitagawa, I. (1984) Effects of soyasaponins on liver injury inducced by highly

peroxidation fat in rats. Planta Med. 1984,440-441.

Ohuchi, K., Watanabe, M., Ozeki, T., and Tsurufuji, S. (1985) Pharmacological

influence of saikosaponins on prostaglandin E2 production by peritoneal

macrophages. Planta Med. 1985,208-212.

Okubo, K. (1993) Soybean saponin: Structure, inheritance, distribution in genus Glycine,

and antiviral activity on HIV and EBV in vitro.

Abstracts of xv Internation Botanic Congress, Yokohama, p. 101.

Osakai, T., Nuno, T., Yamamoto, Y., Saito, A., and Senda, M. (1989)

A microcomputer-controlled system for ion-transfer voltammetry.

Bunseki Kagaku, 38, 479-485.

Peden, D. B., Hohman, R., Brown, M. E., Mason, R. T., Berkebile, C.,

Fales, H. M., and Kaliner, M. A. (1990) Uric acid is a major antioxidant in human

nasal airway secretions. Proc. Natl. Acad. Sci. USA 87, 7638-7642.

Pething, R., Gascoyne, P. R. C., McLaughlin, J. A., and Szent-Gyorgyi, A.

(1983) Ascorbate-quinone interactions: Electrochemical, free radical, and cytotoxic

properties. Proc. Natl. Acad. Sci. USA 80, 129-l32.

Takahashi, M., Niki, E., Kawasaki, A., Kumasaka, A., Yamamoto, Y.,

Kamiya, Y., and Tanaka, K. (1986) Oxidation of lipids. XIV. Inhibition of

oxidation of methyllinoleate by fatty acid esters of L-ascorbic acid.

Bull. Chem. Soc. Jan. 59, 3179-3183.

Tsurumi, S., Takagi, T., and Hashimoto, T. (1991) A new UV -B-absorbing

substance in pea seedling.

Fourth Congress of the European Society for Photobiology, Amsterdam, Abst., p. 1l3.

Tsurumi, S., Takagi, T., and Hashimoto, T. (1992) A y-pyronyl-triterpenoid saponin

from Pisum sativum. Phytochemistry 31, 2435-2438.

74

van der Veen, R., Heimovaara-Dijkstra, S., and Wang, M. (1992) Cytosolic

alkalinization mediated by abscisic acid is necessary, but not sufficient, for abscisic acid­

induced gene expression in barley aleurone protoplasts. Plant Physiol. 100, 699-705.

Wayner, D. D. M., Burton, G. W., Ingold, K. V., Barclay, L. R. C., and

Locke, S. J. (1987) The relative contributions of vitamin E, urate, ascorbate and

proteins to the total peroxyl radical-trapping antioxidant activity of human blood plasma.

Biochirn. Biophys. Acta 924, 408-419.

Williams, N. H. and Yandell, J. K. (1982) Outer-sphere electron-transfer reactions of

ascorbate anions. Aust. J. Chern. 35, 1133-1144.

Wittenberg, B. A. and Wittenberg, J. B. (1985) Oxygen pressure gradients in isolated

cardiac myocytes. J. Bioi. Chern. 260, 6548-6554.

Yamanaka, T. and Kamen, M. D. (1967) An NADP reductase, an NADH-dye reductase

and a non-haem iron protein isolated from a facultative photoheterotroph,

Rhodopseudornonas palustris. Biochirn. Biophy. Acta 131, 317-329.

Zweier, J. L., Flaherty, J. T., and Weisfeldt, M. L. (1987) Direct measurement of

free radical generation following reperfusion of ischemic myocardium.

Proc. Natl. Acad. Sci. USA 84, 1404-1407.

75

Publication List

Tsujino, Y., Tsurumi, S., Osakai, T. and Hashimoto, T. (1993) y-Pyronyl­

triterpenoid Saponin (Chromosaponin I) as an Electron Donor.

Plant Cell Physiol. 34, s62.

Tsujino, Y., Tsurumi, S., Osakai, T., and Saito, A. (1994) A New Triterpenoid

Saponin (Chromosaponin I) with a Reducing Power. Chern. Lett. 1994,711-714.

Tsujino, Y., Osakai, T., and Tsurumi, S. (1994) The Reducing Power of y-Pyronyl­

triterpenoid Saponin (Chromosaponin I) and its pH-dependence.

Biochirn. Biophys. Acta submitted.

Tsujino, Y. and Tsurumi, S., (1994) A New Triterpenoid Saponin, Chromosaponin I,

Scavenges Stable Free Radicals. Plant Physiol. 105, s89.

Tsujino, Y., Tsurumi, S., Yoshida, Y., and Niki, E. (1994) Antioxidative Effects

of Dihydro-y-Pyronyl-Triterpenoid Saponin (Chromosaponin I).

Biosci. Biotech. Biochern. 58, in press.

Tsujino, Y., Tsurumi, S., Yoshida, Y., Gotoh, N., and Niki, E. (1994)

Action of y-Pyronyl-triterpenoid Saponin (Chromosaponin I) as antioxidant.

to be submitted.

Tsurumi, S. and Tsujino, Y. (1994) Chromo saponin I stimulates the growth of

lettuce roots. Physiol. Plant. submitted.

Tsujino, Y., Yokoo, Y., and Sakato, K. (1991) Hair coloring and waving using

oxidases. 1. Soc. Cosrnet. Chern. 42, 273-282.

Yamamoto, K., Tsujino, Y., Saito, A., and Sinohara, H. (1985) Concentrations of

murinoglobulin and a-macroglobulin in the mause serum: variations with age, strain, and

experimental inflammation. Biochern. Int. 10,463-469.

76

International Congress

Tsujino, Y., Osakai, T., and Tsurumi, S., (1993) Reducing Power of r-Pyronyl­

triterpenoid Saponin (Chromosaponin I) and its pH-dependence.

Abstracts o/XV International Botanical Congress, Yokohama, p. 377.

77

Acknowledgments

The author wishes to thank Professor Taisaku Amakawa for his support and

encouragement. Thanks also go to Professor Tohru Hashimoto, Professor Atsuyoshi Saito,

Professor Yoshikiyo Ohji, Dr. Seiji Tsurumi, and Dr. Toshiyuki Osakai of Kobe University;

Professor Kozi Asada and Professor Tokuji Ikeda of Kyoto University; Professor Etsuo Niki of

the University of Tokyo; and Professor Masaru Kimura of Nara Women's University, for their

valuable advice throughout this work, and to all other members of the Laboratory of Physiology,

Department of Biology, and Department of Chemistry, Faculty of Science, for their help,

discussion, and encouragement. The author would like to express his sincere gratitude as well

to Yamahatsu Sangyo Kaisha, Ltd. Finally, the author would like to thank all of his family

members.

78