Plant Physiol. (1984) 75, 414-4200032-0889/84/75/0414/07/$O1 .00/0

Effects of Rotenoids on Isolated Plant MitochondriaReceived for publicaton November 14, 1983 and in revised form February 7, 1984

PATRICK RAVANEL*, MICHEL TIssur, AND ROLAND DOUCELaboratoires de Pharmacognosie (P. R) et de Physiologie Cellulaire Vegetale (M. T., R. D.), Universite Ide Grenoble, Boite postale 68, Saint Martin d'Heres cedex, France

ABSTRACr

The effects of severa rotenoids have been studied on potato ( wmtaberoam L.) tuber and etiolated mung bean (Pkasels arnm Roex)hypocotylsm i Te selective inhibition of mitoch l com-plex I is c tI by several tests: (a) no effect can be observed onexogenousNADH ors teoxidatin; (b) malate oxidatio is inhibitedat pH 7.5; (c) one-third d ease of ADP/O ratio appears during malateoxidation at pH 6.5 or dring a-ketoglutarate, citrate, or pyrunvate oxi-dation at a pH about 7; (d) duing malate oxidtio at pH 6.5, a entinhibition appear which can be maintained by addit of exogeanosoxalonetate; (e) in potato mitochodria, the inhibition of malate oxdda-tiodiappe at pH 6.5 when NAD) is added. Te, a one-thirddecrese of the ADP/O rtio can be me d.

Such a selective nhibitio of complex I is obtined with deguelin,tephrosin, elliptone, OH-12 rotenone, and almost all the rotenoids ex-trcted from Denis roots. lTe presence of the rings A, B, C, D, E seemsto be necessary for the selective inhibition. Opening of the E ring andhydroxylation of the 9 posion (rot-2'-enoic acid) give a rotend deriv-ative with mutste inbitwyactivities on flavoproteins, which are quitecomparbl to those of common flavooids such as kaempferol (Ravaneldta. 1982 Plat Physiol 69: 375-378).

At the present time, almost 200 different compounds isolatedfrom angiosperms have been classified in the isoflavonoid family(10). In contrast with the flavonols which are widely distributedamong the plant kingdom, isoflavonoids can only be found in asmall number of botanical families mainly in the iLuminoseae(Lotoideae and Caesalpinioideae). Among Rosaceae, Moraceae,Amaranthaceae, Iridaceae, and Podocarpaceae, some species areable to synthezise isoflavonoids.

Originating from the CIs isoflavone structure itself, some morecomplex molecules (rotenoids, pterocarpans, coumestans) appearin a very small number of botanical genus. Only rotenoids havebeen studied in this work. The rotenoid molecule (Formula I) iscomposed of four rings A, B, C, D; A, C, D rings characterizethe C,s isoflavone structure; an additional ring E is often presentin rotenoids and associated with ring D. Almost 50 differentrotenoids have been isolated from the genus Derris (10), Tephro-sia, Milletia, Mundulea, Lonchocarpus (17), and Piscidia, whichare tropical plants ofthe Leguminoseae family, growing in Africa,Asia, or America. For several centuries, these plants have beenused to prepare hunting or fishing poisons. More recently, theirinsecticidal properties were discovered in Europe and severalscores of tons of plant dry powder are sold every year at thepresent, as a major constituant ofhouse insecticidal preparations.

Rotenoids are generally located in the superficial parts of theunderground organs of the plants and often reach a very highconcentration. The most common rotenoid is rotenone itself

(Formula I). It has been first isolated from the roots of RobiniaNicou Aublet. This plant is commonly named 'roten' in Taiwan.

7'

ELUPTONE

UGM3 CWHG ROTENONE cu OSAJIN

OH 0 CM2~~OH03 CH OH

OCH3 C

CH3 CH3Formula 1. Structure of the different studied flavonoids.

At present, rotenone is often used in laboratory experimentson animal or plant mitochondria (9, 11) because of its inhibitoryproperties on complex I. In this work, we have compared theactivity of other rotenoids to that ofrotenone itself, on two typesof plant mitochondria in order to obtain some information onthe structural features needed to obtain a spcific inhibition ofcomplex I and to determine the conditions under which therotenoids are efficacious.

MATERIALS AND MEHMODS

Prepari of Miochondria. Mitochondria from potato tu-bers (Solanum tuberosum L) and etiolated mung bean (Phas-eolus aureus Roxb.) hypocotyls cut from bean seedlings grownfrom 5 d in the dark at 26*C and 60% RH. were pred andpurified by methods previously described (8). All operations werecarried out at 0 to 4C. Following purifications, the mitochondriaappeared to be virtually free from extramitochondrial contami-nation and had a high degree ofmembrane intactness, asjudgedby electron microscopy and by low activities of the inner mem-brane and matrix marker enzymes (antimycin A-sensitiveNADH; Cyt c oxidoreductase and malate dehydrogenase). In

414

Dow

nloaded from https://academ

ic.oup.com/plphys/article/75/2/414/6079731 by guest on 19 N

ovember 2021

ROTENOID EFFECTS ON PLANT MITOCHONDRIA

addition, the mitochondria were tightly coupled: averaged ADP/0 ratio for succinate was 1.8 and respiratory control ratio for thesame substrate was about 3.02 Uptake Measurements. 02 uptake was followed polaro-

graphically at 25°C using a Clark-type electrode system purchasedfrom Hansatech Ltd. (Hardwick Industrial Estate, Kings Lynn,U.K.). The reaction medium contained 0.3 M mannitol, 5 mMMgC2, 10 mm KCl, 10 mm Pi buffer, 0.1% defatted BSA(medium A), and known amounts of mitochondrial proteins.Unless otherwise stated, all incubations were carried out at pH7.2.Uncoupling Test. An amount of intact mitochondria corre-

sponding to 0.30 mg protein was suspended in the electrodemedium containing 6 mm succinate and 300 Mm ATP. After astate HII-state IV transition, 10 Mm carboxyatractyloside wasadded in order to inhibit the nucleotide carrier. The uncouplingeffect ofa substance added at this stage corresponds to an increaseofthe oxidation rate; 100% uncoupling effect was obtained whenthe rate of 02 consumption was not further stimulated by theaddition of FCCP' (1 AM). After each assay, the incubationmedium was rapidly centrifuged at l0,OOOg for 3 min (BeckmanMicrofuge B).

Intact mitochondria were suspended in reaction medium(NH4Cl or NH4NO3 150 mM, Tris HCI 10 mM, BSA 0.1%, pH7.2). Passive swelling reactions were measured by changes at 540nm in a Beckman spectrophotometer (model 25) as previouslydescribed (13). Potato mitochondria show a slow rate of swellingin NH,+ salts. In these conditions a rapid passive swelling isinduced by uncouplers.

Chemicals. Rotenone was purchased from Sigma. All otherproducts were a generous gift from F. Delle Monache (Institutodi Chimica, Roma) for tephrosin, deguelin, OH-12 rotenone,and P. B. Anzeveno (Dow Chemical Laboratories) for elliptone,deguelin, rot-2'-enoic acid. All these products were dissolved inethanol. The concentration of ethanol in the reaction mediumnever exceeded 3%. At this concentration, ethanol alone isalmost without effect on mitochondrial respiration under ourconditions (presence of BSA). Usually, 300 AM was the upperconcentration of rotenoids used in our experiments. For greaterconcentrations, these compounds tend generally to precipitateout from the reaction medium. Rotenoid fractions of Derris rootdry powder were obtained by TLC. Powder was submitted toethanolic extraction in the dark. The crude extract was chromat-ographed in an atmosphere devoid of02, on 0.5-mm thick layersof silica gel GF 254 (Merck). The solvent used for TLC waspetroleum ether, chloroform, acetone (4:1:1, v/v). Chromato-grams were observed under shortwave (254 nm)UV light. TwelveUV absorbing bands were detected with increasing RF from themost hydrophilic band 12 (RF = 0.1) to the most lipophilic bandI (RF = 1). The bands were extracted with a small amount ofethanol and rechromatographed. After different purifications,the extract was concentrated to 200 ul. An aliquot ofeach fractionwas cochromatographed in the presence of authentic samples indifferent solvent systems: dichloromethane, acetone (95:5, v/v)or chloroform, acetone, acetic acid (196:3:1, v/v) or petroleumether, chloroform, acetone (4:1:1, v/v). Spectra of fracions wereperformed in ethanol and compared to spectra of authenticsamples. Concentrations were estimated with the e of rotenone(X max = 290 nm). We have verified that all these compoundswere not oxidized or altered by the mitochondria.

I Abbreviations: FCCP, carbonyl cyanide p-trifluoromethoxy-phenylhydrazone; MDH, malate dehydrogenase; ME, malic enzyme;SHAM, salicylhydroxamic acid; TPP, thiamine pyrophosphate.

RESULTS

Deguelin Effects on Plant Mitochondrial Activities. (a) Cou-pling between Electron Transfer and Oxidative Phosphorylation.At pH 7.2, with 6 mm succinate as substrate, deguelin at concen-tration between 1 and 300 Mm (solubility limit) does not induceany change either on the respiratory control or on the ADP/0ratio which value remains close to 1.8 (Fig. 1). Moreover, degue-lin at the same concentrations, fails to show any uncouplingproperties when an uncoupling test, descried before (25), isemployed.

In the same way, deguelin at concentrations between 1 and300 Mm does not induce sweling ofmitochondria in the presenceof NH.4C1 or NH4NO3 0.15 M (result not shown); deguelin is,therefore, unable to allow a transfer of protons through themitochondrial inner membrane.

All these results together demonstrate that deguelin is ineffec-tive on the phosphorylation mechanism, i.e. proton gradientformation, activities of the coupling factor, and of nucleotidecarrier.

(b) Deguelin Effects on the Electron Transfer with Succinate orNADH (Fig. 2). Deguelin at concentrations between 1 and 300AM iS without effect on the rates ofsuccinate orNADH oxidationin the presence of ADP or of an uncoupler such as FCCP, inpotato or mung bean mitochondria. Deguelin is, therefore, with-out inhibitory properties either on the external NADH dehy-drogenase, succinate dehydrogenase, and complex III and IV.

(c) Deguelin Effect on Malate Oxidation (Fig. 2). Malateoxidation by plant mitochondria involves the enzymeMDH andNAD+-linked ME, both ofwhich are loclized in the mitochon-drial matrix (7). ME catalyses the oxidative decarboxylation ofmalate to pyruvate with NAD+ and Mn2+ as cofactors (15). Theactivity ofME in intact mitochondria is influenced by a numberof factors, including the pH of the external medium (16). Athigher pH (7.5 or more depending on the species), ME is lessactive than at lower pH. The concentration ofNAD+ within themitochondrial matrix can also influence the activity ofME, withhigher concentrations stimulating, and this can be altered byadding NADI to isolated mitochondria (27).MDH activity is reversible and, in the mitochondrial matrix,

the characteristics of malate oxidation strongly support the evi-dence that the regulation in vivo ofMDH activity can be readilyaccounted for by equilibrium effect alone (2, 20). MDH activity,on the other hand, is more active at higher pH. When oxaloac-etate and NADH concentrations increase in the matrix space,MDH produces malate; MDH catalyzes the formation of oxal-oacetate and the reduction ofNAD+ only when the ratios oxal-oacetate/malate and NADH/NAD+ are low. At pH 7.5, MDHis operating alone (2, 20). Under state m conditions, NADH/NADI ratio is low and the equilibrium ofthe MDH reaction hasmoved toward oxaloacetate formation. One of the reasons whypotato mitochondria are able to consume 02 at significant ratesat high pH with malate as substrate in the absence ofa biochem-ical system to remove oxaloacetate, is because they excreteoxaloacetate into the external medium (20).At pH 7.5 in the presence of 5 gM deguelin, 02 uptake is

inhibited (Fig. 2). This inhibition could concern complex I orMDH itself. Mitochondrial treatment with Triton X100 permitsto obtain a MDH solution. The formation of NAD+ by theextracted enzyme, measured spectrophotometrically, in the pres-ence of oxaloacetate and NADH is not modified by addition ofdeguelin between 1 and 300;tM (result not shown). Since deguelinis without effect on NADH and succinate oxidation, the inhibi-tory activity of deguelin observed with malate is, therefore,located at the level of complex I. As for rotenone, the strongeffect ofdeguelin can be explained as an inhibition ofthe electrontransfer at the level of one of the several non-haem iron centersassociated with the internal NADH dehydrogenase complex I

415

Dow

nloaded from https://academ

ic.oup.com/plphys/article/75/2/414/6079731 by guest on 19 N

ovember 2021

Plant Physiol. Vol. 75, 1984

ATP SuccinateM4 ADPIOOPM

170arvbmcytrWcyloudd R3.

50\ ADPip100p %: .7so 1pMDW 0p

IoMMM D

I\g pM D

I I P2olM D

20pM 02 300p DI YFCCPl ipM

-. 1 min . 230

IAAA8 8

(5). As a consequence, the internal NADH/NAD+ ratio is raisedand the equilibrium of the malate dehydrogenase reaction isdisplaced toward malate formation.At pH 6.5, ME is strongly active. Addition of 1 juM deguelin

induces a transient inhibition of state III rate of 02 uptake withmalate as substrate (Fig. 2). At the same time, the ADP/O value,which was near 3 with malate before addition of deguelin,decreases by one-third (Table I) indicating that the deguelin-resiant pathway is nonphosphorylating. These results suggestthat, at pH 6.5 in the presence of 1 gM deguelin, as with rotenone,the complex I which possesses a hig affinity for NADH, isinhibited and is bypassed by another NADH dehydrogenase,which is rotenone resistant, nonphosphorylating, and has a muchlower affinity for NADH (3). The transient inhibition which

RC= 3.4P/- 1.780

FIG. 1. Effect ofdeguelin on succinateoxidation by potato tuber mitochon-dria. The oxidation of 6 mM succDatewas measured as 02 consumption inmedium A (see "Material and Meth-ods") with 0.30 mg mitochondrial pro-teins m1l-, in the presence of 150 gMATP. 10 M carboxyatractyloside and 5pM antimycin A (A.A) were addedwhere indicated Dotted lines show thereference trace after uncoupling with 1ptM FCCP. Numbers on traes refer tonmol 02 consumed min-' mg7' protein.

FIG. 2. Effect of deguelin on state HI ratesof substrate oxidations by potato tuber mito-chondria. The oxidations of substrates (1 mMNADH, 6 mM succinate + 150 FM ATP, 10mm citrate, 10 mM a-ketoglutarate + 150 pMTPP + 2 mM malonate, 15 mM malate) weremeasured as 02 consumption in medium A(see "Material and Methods") with 0.30 mgmitochondrial protein ml1', ADP (2 mM), de-guelin (D), and antimycin A: A.A. (5 pM) wereadded where indicated. Numbers on traces re-fer to nmol 02 consumed min-' mg-' protein.

appears just after injection of deguelin reflects the concertedoperation of MDH and ME (23). Just after the addition ofdeguelin, previously accumulated oxaloacetate reenters the mi-tochondria and is reduced via MDH at the expense of NADHproduce by ME; consequently, 02 uptake is initially slow. Whenall the oxaloacetate has been removed, NADH becomes accessi-ble to the respiratory chain again and 02 uptake rates increase(Fig. 2). Under these conditions, the rotenone-insensitive internalNADH dehydrogenase may be engag (18). This inetationofbiphasic inhibition by deguelin is supported by analysis oftheproducts of malate oxidation which has shown that pyruvateaccumulates while oxaloacetate levels decrease during the initialphase of inhibition (result not shown). It is interesting to notethat these results obtained with deguelin atpH 6.5 concern potato

416 RAVANEL ET AL.

'SuWnate

S'YADPSOPMM 3.4

SOPM % - 1.78

kDPSOpM

'-.,3000M D

Dow

nloaded from https://academ

ic.oup.com/plphys/article/75/2/414/6079731 by guest on 19 N

ovember 2021

ROTENOID EFFECTS ON PLANT MITOCHONDRIA

Table I. Effect ofDeguelin on the ADP/O Ratio Values ObservedDuring the Course ofSubstrate Oxidation by Potato Tuber

MitochondriaSubstrates: 1 mm NADH, 15 mm malate + 1 mM NAD+, 10 mM a-

ketoglutarate + 150 iM TPP + 1 mM NAD4, 5 mMpyruvate + 0.5 mMmalate + 150 AM TPP + 1 mm NAD4; in these conditions, the rte ofpyruvate oxidation remains sufficient (i.e. oxaloacetate delivery is suffi-cient), after 10 pM deguelin addition, to permit a good determination ofADP/O.

ADP/0 ADP/0 after 10 pMDeguelin Addition

NADH 1.6 ± 0.15 1.6Malate (pH 6.5) 2.6 ± 0.15 1.3a-Ketoglutarate 3.6 ± 0.20 2.7Pyruvate 2.4 ± 0.15 1.3

z0

z

0 AAH

1 & 4 5//100 200300

DEGUELIN CONCENTRATION (pM)

FIG. 3. Effect of deguelin on state III rates of substrate oxidation byNAD+-deficient potato mitochondria. Substrates: 1 mM NADH (pH 7.2,I = 250), 6 mM succinate + 150 uM ATP (pH 7.2, I = 185), 10mM citrate(pH 7.2, I = 32), 10mM a-ketoglutarate + 150 AM TPP + 2 mM malonate(pH 7.2, 1 = 40), 5 mM pyruvate + 0.5 mM malate + 150 AM TPP (pH7.2, I = 37) 15 mM malate (pH 6.5, I = 29). I = Initial oxidation rate(nmol 02 consumed min-' mg7' protein). When deguelin concentrationin the medium is 1 AiM it corresponds to 1 nmol mg' protein.

mitochondria only after addition of exogenous NAD+ or mungbean mitochondria with or without NAD+ (27). This can beexplained by the fact that, in contrast with mung bean mito-chondria, potato mitochondria are very often depleted of theirNAD+ content. Addition of exogenous NAD+ is followed by itsselective transport through the inner membrane (27) and allowsa high level of ME activity. Without added NAD+, internalNADH concentration is not sufficient to engage the rotenone-insensitive internal NADH dehydrogenase and deguelin is thenactive as inhibitor at pH 6.5 as at pH 7.5 (Fig. 3). In addition,Figure 2 shows that, in the presence ofdeguelin, malate oxidationat pH 6.5 is strongly inhibited by antimycin A.

(d) Deguelin Effects on Citrate and a-Ketoglutarate Oxidation.Figure 2 shows the oxidation of citrate and a-ketoglutarate atpH 7.2 and in the presence ofNAD+ (i.e. when the matrix NAD+concentration is high). After addition of 1 jiM to 300 uM deguelin,no apparent inhibition ofthe 02 consumption appears, butADP/0 decreases 1 unit with either citrate or a-ketoglutarate in thepresence of 0.3 mm malonate and 0.3 mMTPP (Table I). Whenendogenous NAD+ concentration is too low, (i.e. potato tubermitochondria in the absence ofexternal NADI), 02 consumption

is strongly inhibited by deguelin (Fig. 3). When NADI is addedto the reaction medium, 02 consumption rate with a-ketoglutar-ate or citrate at pH 7.2, in the presence ofdeguelin, is comparableto 02 consumption obtained with malate at pH 6.5 (see Fig. 2).Nevertheless, the transient inhibition never appears (Fig. 2) witha-ketoglutarate or citrate, because no oxaloacetate is accumu-lated in the medium.

Effects of Rotenoids on Plant Mitochondrial Activities. Fivedifferent rotenoids have been compared: deguelin, tephrosin,elliptone, 12-OH rotenone, and rot-2'-enoic acid (Formula I).Rot-2'-enoic acid is a structure appearng in rotenoid intercon-version (1). The four first above mentioned rotenoids are withouteffect on 02 consumption by potato or mung bean mitochondriawith NADH or succinate as substrates. On the contrary, rot-2'-enoic acid induces an inhibition of the electron flow with thesesubstrates.

Figure 4 (A and B) shows that each of the studied rotenoidsseverely inhibits 02 consumption when malate is the substrate atpH 7.5. A 100% inhibition is obtained with 2 to 5 pM deguelin;however, rotenone, OH-12 rotenone, tephrosin, and elliptoneare slightly less potent inhibitors, either with mung bean hypo-cotyl or potato tuber mitochondria. Rot-2'-enoic acid is muchless effective than the other rotenoids; a 100% inhibition isobtained at a concentration higher than 300 pM.

In mitochondria where endogenous NADI concentration isnot a limiting factor (mung bean mitochondria or potato mito-chondria supplemented with exogenous NAD+), at pH 6.5 withmalate as substrate, each of the studied rotenoids induces atransient inhibition of state m rate of 02 uptake (Fig. 5). Again,after addition of each of the studied rotenoids (except for rot-2'-enoic acid), ADP/O ratio values decrease 1 unit (Table II). Therotenoid insensitive pathway which is operating at this stage canbe fully inhibited by antimycin A. When endogenous NAD+concentration is too low-aged potato mitochondria (2 1-atpH 6.5 with malate as substrate, all of these rotenoids stronglyinhibit 02 uptake, as at pH 7.5. Whereas rotenone, elliptone,deguelin, tephrosin, and OH-12 rotenone appear to be quitecomparable selective inhibitors of complex I, rot-2'-enoic acidhas a particular behaviour. its inhibitory efficiency on complexI is lower, 100 pM are needed for a 70 to 85% inhibition ofmalate oxidation at pH 7.5. Moreover, it has a non-negligibleinhibitory action on NADH or succinate oxidation (40 and 25%,respectively at saturation). In addition, this compound, in con-trast with the other studied rotenoids, exhibits an uncouplingeffect. This effect appears at a concentration of 250 gM. Itsintensity represents 95% ofthe FCCP effect in potato mitochon-dna and 90% in mung bean mitochondria. As a whole, rot-2'-enoic acid can be classified among the inhibitory uncouplers(25).

Effects of Different Flavonoids Isolated from Derris RootPowder on Mitochondrial Activities. It was interesting to com-pare the pure rotenoids studied here with the whole flavonoidmixture present in the roots ofone ofthe best source ofrotenoids,the genus Derris. For this purpose, Derris root dry powder wassubmitted to ethanolic extrction in the dark, at laboratorytemperature, in an atmosphere devoid of O2. Twelve differentUV2so absorbing bands were obtained after TLC and were puri-fied by rechromatography with the same procedure. TLC indifferent solvents in the presence of test substances and UV-visible spectral analysis allowed the identification of rotenoneand deguelin in the band 6, of elliptone (band 7) and of 12-OHrotenone (band 8). Tephrosin seemed not to be present in thestudied extract of Derris root powder. After purification, thecompounds isolated from each ofthe 12 bands were dissolved inethanol, submitted to spectrophotometric quantitative estima-tion and their effects on plant mitochondria were studied.At concentrations lower than 50 gM none of the studied

417

Dow

nloaded from https://academ

ic.oup.com/plphys/article/75/2/414/6079731 by guest on 19 N

ovember 2021

Plant Physiol. Vol. 75, 1984

top

FIG. 4. Effects ofvarious rotenoids on malate oxidation by potato tuber (A) and mung bean hypocotyls (B) mitochondria. Dotted lines show theinhibition with the rotenoids reference: rotenone. Substrate: 15 mM malate + 1 mM NAD+ (pH 7.5).

compounds inhibits NADH or succinate oxidation. On the con-trary, every compound strongly inhibits malate oxidation at pH7.5 (Fig. 6). This inhibition is greater on potato than on mungbean mitochondria. 150 (concentration giving a 50% inhibitionof the electron flow) measurements at pH 7.5 with malate assubstrate allows a classification of the different compounds: (a)compounds corresponding to bands 5, 6, 8 give an 150 between0.3 and 1.5 jM. These compounds are as good or better inhibitorsas rotenone itself. (b) Compounds coresonding to bands 1, 3,9, 7 give an Iso between 1.5 and 4.5 AM; (c) compounds corre-sponding to bands 4, 10, 11, 12 give an Iso between 5 and 10 lM.When the lipophilic character of these flavonoids decrases(bands 10, 11, 12) their inhibitory potency decreases sightly. AtpH 6.5, with malate as substrate in the presence of nonlimitingconcentration ofNADI, the compounds corresponding to bands5, 6, 7, 8, 9 induce a transient inhibition which is followed by amarked increase ofthe 02 consumption. At this stage, a decreaseof I unit of the ADP/O ratio can be shown (Table II).

DISCUSSION

The experimental procedures employed here permit the char-actrization of products inhibiting selectively complex I. Thesetests derive from about 30 years of scientific rch which hasprogresavely explained the mode of action of compounds suchas rotenone, barbiturates, amytal, or piericidin A on plant oranimal mitochondria.

In plant mitochondria from several botanical species, thespecific inhibition of maoate ion by rotenone has beenstudied by Chauveau and Lance (6), Palmer (22), Wilson andHanson (28), Wiish and Day (29). Our results demonstratethat under state HI conditions deguedin and various rotenoidsstrongly inhibit the coxidation of all NAD+-linked substates

insofar as the total amount of mitochondrial NAD+ is low. Inthe presence of NAD+, only the oxidation of malate in state IIIis rotenone sensitive. The degree of malate inhibition inducedby deguelin is also linked to the concentration of oxaloacetate.

ROENOIDE CONCENTRATION (PM)

418 RAVANEL ET AL.

Dow

nloaded from https://academ

ic.oup.com/plphys/article/75/2/414/6079731 by guest on 19 N

ovember 2021

ROTENOID EFFECTS ON PLANT MITOCHONDRIA

FIG. 5. Effect of various rotenoids on stateIII rate of malate oxidation by potato tubermitochondria. The oxidation of 15 mM malate+ 1 mM NAD, pH 6.5 was measured as 02consumption in medium A (see "Material andMethods") with 0.30 mg mitochondrial proteinml-'. ADP (1 mm), rotenoids, and antimycinA (5 ,M) were added where indicated. Numberson taces refer to nmol 02 consumed/min *mgprotein.

Table IL Effect of Various Rotenoids on theADP/O Ratio Values Observed During the Course ofMalateOxidation by Potato Tuber and Mung Bean Hypocotyl Mitochondria

Substates: 15 mM malate + 1 mm NAD+ (pH 6.5).

Concentration(AM)

Potato Mung Bean

ADP'O ADP/O after ADP/O ADP/O afterrotenoid addition rotenoid addition

Rotenone 5 2.7 1.5 2.6 1.4Deguelin 2 2.6 1.6 2.6 1.512-OH rotenone 20 2.4 1.3 2.3 1.3Tephrosin 50 2.4 1.7 2.5 1.4Elliptone 50 2.8 1.7 2.8 1.7Rot-2'-enoic acid 50 2.7 0.8 2.7 0.9TLC band 5 5 2.3 1.4TLC band 6 5 2.3 1.3TLC band 7 10 2.3 1.2TLC band 8 10 2.3 1.3TLC band 9 10 2.3 1.3

It has previously been observed that the ADP/O ratio withNADI-linked substrates was decreased by one-third in the pres-ence of rotenone (3). The same thing holds with deguelin andvarious rotenoids (Table II). These results indicate that thedeguelin-reistant pathway is nonphosphorylating and that de-guelin like rotenoids engage the functioning of an internal rote-none-insensitive NADH dehydrogenase which is only activeunder conditions where the matrix concentration of NADH ishig (i.e. in mung bean hypocotyl mitochondria or NAD+-supplemented potato tuber mitochondria).The results obtained with rotenone have led us to define better

the tests needed to assess that a compound is a selective inhibitorofcomplex I: (a) no effect on NADH or succinate oxidation; (b)inhibition of malate oxidation at pH 7.5; (c) one-third decreaseof the ADP/O ratio during malate oxidation at pH 6.5 or a-ketoglutarate, citrate or pyruvate oxidation at a pH about 7; (d)during malate oxidation at pH 6.5, transient inhibition whichcan be maintained by addition of exogenous oxaloacetate; (e) inNADI-deficient mitochondria disappearance ofthe inhibition ofmalate oxidation at pH 6.5 when NAD+ is added. Then a one-third decrease of the ADP/O can be measured.We have shown that all the rotenoids tested here (with the

exception of rot-2'-enoic acid) are acng in the same way andare, therefore, selective inhibitors ofthe complex I. Deguelin andprobably the compound corresponding to band 8 from Derrisroots are more potent inhibitors than rotenone itself. When thering A, B, C, D, E of the rotenoid structure are present in theformula, structural details concerning the E ring seem not tochange the selective inhibition of complex I. Furthermore, the6',8' double bond present in rotenone itself does not play animportant role for the inhibitory potency. In contrast with John-son-Flanagan and Spencer (12), the effects ofelliptone are similarto those observed with rotenone and we have never observed anyconversion of rotenone to elliptone in the reaction medium.Replacement of-0 by-OH on the 12 position or suppressionof the lateral chains on the 5' position induce a slight decreaseof the selective inhibitory activity. A comparable result is ob-tained when a -OH is added on the 12-a position. In markedcontrast, opening of the E ring induces a great loss of theinhibitory potency (case of the rot-2'-enoic acid). In the sameway, suppression of the B ring (case of the isoflavone osajin:formula 1, result not shown) gives products without selectiveinhibitory action on complex I. On the whole, our results showthat the presence of rings A, B, C, D, and E is necessary to obtain

419

Dow

nloaded from https://academ

ic.oup.com/plphys/article/75/2/414/6079731 by guest on 19 N

ovember 2021

RAVANELETAL.~~~~~PlantPhysiol. Vol. 75, 1984

FIG. 6. Concentrations of different compounds (TLC fractions,

text) extracted from Derris root powder giving a 50% inhibition (0o)onstate III rates of malate oxidation by potato tuber and mung

hypocotyl mitochondria. Substrate: 15 mM malate + 1 mM NAD+; mMADP (pH 7.5).

the selective inhibition of complexI. it seems that all rotenoids

presenting these five rings in their structure will be selective

inhibitors of complexI, with a more or less good efficiency,

following the substitutions. We are not able to estimate the

played by substitutions on ring A because all the analyzed

rotenoids were substituted in the same way by -OCH3 on the

second and third positions. It is interesting to compare our results

with those of Burgos and Redfearn (4) on the effect of various

rotenoids on isolated complex I from rat liver mitochondria.

They demonstrate that without the E ring, or when the B or C

rings are open, rotenoids are weakly active on purified complex

I. The complex I of animal mitochondria seems to be much

more sensitive to rotenone than that of plant mitochondria.

Our work on rot-2'-enoic acid shows a loss of selective activity

on complexI. This compound has inhibitory properties on the

mitochondrial electron transfer which seems to be located in the

flavoprotein region. Such medium multisite inhibition can be

obtained with a great number ofother flavonoids like kaempferol

(26) or with a great variety of other compounds. Such effects

could perhaps be explained by a loss of inner membrane fluidity,

induced by all these substances (19). Nevertheless, we have shown

(14, 24, 26) that such multisite inhibitions do not have exactly

the same characters according to the different compounds tested.

It is interesting to note that uncouplingpropertes appear in the

rotenoid group only with rot-2'-enoic acid. Opening of the E

ring and hydroxylation of the 9 position seem to participatethe induction of this newproperty. Free hydroxyls present

tephrosin and OH-12 rotenone are unable to permit uncoupling.Our results show that, in Derris root, not rotenone alone but

all the complex pattern of rotenoids is active on mitochondrial

complexI. These rotenoids are present in the plant as free

aglycones, reaching very high concentrations in the root. It is,

therefore, possible that the regions accumulating rotenoids could

beanatomically or physiologically isolated from the cytoplasm.In the Derris root, the sinificance of such ahigh rotenoid

accumulation seems evidently to be ecophysiological. Rotenoidsare certainly toxins preventing predation or parasitism on theroot by animals, espially invertebrates.

LITERATURE CITED

1. ANZEVENO PB 1979 Rotenoid interconversion. Synthesis of elliptone fromrotenone. J. Heterocycl chem 16: 1643-1644

2. BOWMAN EJ, H IKUMA 1976 Regulation of malate oxidation in isolated mungbean mitochondria. II. Role of adenylites Plant Physiol 58: 438-446

3. BRUNTON CJ, JM PALMER 1973 Pathways for the oxidation of malate andreduced pyridine nucleotide by wheat mitochondria. Eur J Biochem 39:283-291

4. BuRGOs J, ER REDFEARN 1965 The inhibition of mitochodrial reduced nico-tinamide adenine dinucleotide oxidation by rotenoids. Biochim BiophysActa 110: 475-483

5. CHANCE B, L ERNSTER, PB GARLAND, CP LEE, PA LIGHT, T OHNIsHI, CIRAGAN, DWONG 1967 Flavoproteinsof the mitochondrial respiratorychain.Proc Natl Acad Sci USA 57: 1498-1505

6. CHAUVEAU M, C LANCE 1971 Les mitochondries de l'inflorescence de chou-fleur. II. Action des inhibiteurs de transport deselectrons. Physiol Veg 9:353-372

7. DAY DA,GP ARRON, GG LATIES 1979 Enzyme distribution in potato mito-chondria. J Exp Bot 30: 539-549

8. DOUCE R, EL CHRsmEN, WD BONNER 1972 Preparation of intact plantmitochondria. Biochim Biophys Acta 275: 148-160

9. ERNSTER L, G DALLNER, GF AZZONE 1963 Differential effects of rotenone andamytal on mitochondrial electron and energy transfer. J Biol Chem 238:1124-1131

10. HARDORNE JB 1971 In JB Harborne, D Boulter, BL Turner, eds, Chemotax-onomy of theL-guminosae. Academic Press, London, pp 31-71

1 1. IKuiA H, WD BONNER 1967 Properties of higher plant mitochondria. III.Effects of r inhibitos Plant Physiol 42:1535-1544

12. JOHNSON-FLANAGAN AM, MS SPENCER 1981 The effect of rotenone on respi-ration in pea cotyledonmitochondria Plant Physiol 68:1211-1217

13. JUNG DW,GP BIERuLY 1979 Swelling and contraction of potato mitochon-dna. Plant Physiol 64: 948-953

14. MACHERE D, P RAVANEL, M Tissur 1982 Effects of herbicidal carbamateson mitochondria and chloroplast Pestic Biochem Physiol 18: 280-288

15. MACRAE AR, R MooRHousE 1970 The oxidation of malate by mitochondriaisolated from cauliflower buds. Eur J Biochem 16: 96-102

16. MACRAE AR 1971 Effect of pH on the oxidation of malate by isolatedcauliflower bud mitochondria. Phytochemistry 10: 1453-1458

17. MENICHINI F, F DELLE MONACHE, GB MAmN-BBmoLo 1982 Flavonoids androtenoids from Tephrosieae and related tribes of Leguminosae. Planta Med45: 243-244

18. MOLLER IM, JM PALMER 1982 Direct evidence for the presence of a rotenonerestant NADH dehydrogenase on the inner surface of the inner membraneof plant mitochondria. Physiol Plant 54: 267-274

19. MORELAND DE, SC HUBER 1979Inhibition of photosynthesis and respirationby subtituted 2-6 diniraniline herbicides.m. Effects on electrontrasportand membrane properties of isolated mung bean mitochondria. Pestic Bio-chem Physiol 110: 247-257

20. NEUBURGER M, R Douce 1980 Effect of icoe andoxaoetate onmalate oxidation by spinach leafmi Bichim Biophys Acta 589:176-189

21. NEUBURGER M, RDOUCE 1983 Slow passive diffusion ofNAD between intactisolated plant mitochondria andsuspening medium. Biochem J 216: 443-450

22. PALMR JM 1976 Theoranization and r ln ofelctrontransport inplant mitochondria. Annu Rev Plant Physiol 27: 133-157

23. PALMER JM,GP ARRON 1976 The influence of exogenousnicotinamideadenine dinucleotide on the oxidation of malate by Jerusalem artichokemitochondria. J Exp Bot 27: 418-430

24. RAVANEL P, M TiSSUr, R DouCE 1981 Effects of flavone on the oxidativeproperties of intact plant mitochondria. Phytochemistry 20:2101-2103

25. RAVANELP, MTWsur, R DouCE 1982 Uncoupling activities of chalcones anddihydrochalcones on isolatedmitochondria from potato tubers and mungbean hypocotyls. Phytochemistry 21:2845-2850

26. RAVANEL P, M TissuT, RDOuCE 1982 Effects ofkaempferol on the oxidativepropertes of intactplant mitochondria Plant Physiol 69: 375-378

27.TOBiN A, BDJERDJOUR, EJOuRNEr, MNEBtURGER, R DoucE 1980 Effect ofNAD+ onmalate oxidation in intact plant mitochondria.Plant Physiol 66:225-229

28.WII50 N RH, JB HANSON 1969The effect ofrespiratoryinhibition NADH,succinate and malate oxidation in corn mitochondria. Plant Physiol 44:1335-1341

29. WsKsH JT, DA DAY 1982 Malateoxidaton, rotenone resitance and alterna-tive path activity in plant mitochondria. Plant Physiol 70: 959-964

420

150(PM)10 b

9-

POTATO M BGEAN8-

7-

6-

5-

4-

3-

2-

1 2 3 4 5 6 7 8 91IDt11 12 1 2 3 4 56 7 8 90 12TLC FRACTIONS

RAVANEL ET AL.

Dow

nloaded from https://academ

ic.oup.com/plphys/article/75/2/414/6079731 by guest on 19 N

ovember 2021