Transmembrane Electron Transport in Plasma Membrane Vesicles Loaded with an NADH-Generating System or Ascorbate

http://www.diva-portal.org

This is the published version of a paper published in Plant Physiology.

Citation for the original published paper (version of record): Askerlund, P., Larsson, C. (1991)

Transmembrane Electron Transport in Plasma Membrane Vesicles Loaded with an NADH-Generating System or Ascorbate.

Plant Physiology, 96(4): 1178-1184

Access to the published version may require subscription. N.B. When citing this work, cite the original published paper.

Free PMC Article

Permanent link to this version:

Transmembrane Electron Transport

in

Plasma

Membrane

Vesicles Loaded

with

an

NADH-Generating

System

or

Ascorbatel

PerAskerlund2 and Christer Larsson*

Department

of Plant

Biochemistry, University

of

Lund,

P.O. Box

7007,

S-220

07

Lund,

Sweden

ABSTRACT

Sugar beet (Betavulgaris L.)leafplasmamembrane vesicles wereloaded withanNADH-generatingsystem(orwithascorbate) and weretestedspectrophotometrically fortheirabilitytoreduce external, membrane-impermeableelectron acceptors. Either al-coholdehydrogenase plus NAD*or100 millimolar ascorbatewas

included in thehomogenizationmedium,andright-side-out (apo-plasticside-out)plasmamembranevesiclesweresubsequently prepared using two-phase partitioning. Addition of ethanol to plasma membrane vesicles loaded with the NADH-generating system led to a production of NADH inside the vesicles which could be recorded at340nanometers. This systemwasable to

reduce 2,6-dichlorophenolindophenol-3'-sulfonate (DCIP-sulfo-nate), astronglyhydrophilicelectron acceptor. The reduction of DCIP-sulfonate wasstimulated severalfold bythe K+ionophore valinomycin, included to abolish membrane potential (outside negative) generated by electrogenic transmembrane electron flow. Fe3+-chelates, such as ferricyanide and ferric citrate, as wellascytochromec, were notreducedbyvesicles loaded with theNADH-generatingsystem. Incontrast, right-side-outplasma membranevesicles loaded with ascorbatesupportedthe reduc-tion of both ferric citrate and DCIP-sulfonate, suggesting that ascorbate also may serve as electron donor for transplasma membrane electrontransport. Differencesinsubstratespecificity andinhibitorsensitivity indicatethat the electrons from ascorbate and NADH were channelled to external acceptors via different electrontransport chains.Transplasmamembrane electron

trans-portconstitutedonlyabout 10% of totalplasmamembrane elec-tron transport activity, but should still be sufficient to be of physiological significance in, e.g. reduction of Fea3 to Fe2+ for uptake.

It has been suggested that the plant plasma membrane contains redox systems that transfer electrons from cyto-plasmic donorstoacceptorsintheapoplast. One such tran-splasmamembrane electrontransport systemisthoughttobe inducedinroots ofnongraminaceous plants duringiron de-ficiency andtobeinvolved in the reduction of

Fe3"

to

Fe2"

foruptake (reviewed in ref. 17). Other, constitutive systems

are thought to be involved in other processes, such as hor-'Supported byagrant from theSwedish Natural Science Research

Council. Part of this workwas presented atthe 8th International

WorkshoponPlant MembraneTransport, Venice, 1989.

2Present address: Department of Plant Sciences, University of

Oxford,South ParksRoad, Oxford,OXI 3RB,U.K.

monal regulation of cell growth or in keeping sulthydryl

groups of membraneproteinsin areduced state (reviewed in refs. 11 and 17). The nature of the electron donor(s) for the transplasma membrane electron transport system(s) is not known. Substantial evidence indicates, however, that NAD(P)H is such anelectron donor (9, 11, 17).

We (1) previously found that

NADH-ferricyanide3

and

NADH-Cytcreductaseactivitieswereabout30%latent (i.e. could bemeasured only in the presence of adetergent, such

as Triton X-100) with inside-out vesicles and about 80% latent with right-side-out vesicles. From these results, we concluded that both donor and acceptorsites for these activi-ties were present on the cytoplasmic surface ofthe plasma

membrane and that transplasma membrane electron trans-port from NADH to ferricyanide or Cyt c could at most constitute 30% of the total activities (1). Based on trypsin

inhibition of NADH-Cyt c reductase activity (2, 21),

H+-pumping capacity (21), and H+-ATPase latency using the detergent Brij 58 (22), we (21) recently estimated the cross-contamination between the inside-out and right-side-out plasma membrane fractions to be about 20%. Hence, the

majorpart (20 of the 30%) of the latentNADH-(acceptor)

reductaseactivities obtained with inside-outvesicles seems to be due to contaminationbyright-side-out vesicles ratherthan to transplasma membrane electron transport. This leaves about 10% of the totalelectrontransportactivityaspossibly transmembrane.

It maybe argued that electronsaretransferred fromNADH acrosstheplasmamembrane to,for example,ferricyanideon the apoplastic side only in the absence ofacceptors on the

cytoplasmicside, that is, thata transportof electronsinthe

planeof the membrane is favoredcompared to transplasma membrane electron transport when both alternatives are pos-sible. In the presentwork,we havetherefore loadedplasma

membrane vesicles with an NADH-generating system and followed the reduction of external membrane-impermeable

electronacceptorstoobtainconditionsthat allowonly trans-membraneelectrontransport. Similarly,wehave also loaded

vesicles withascorbate to see if ascorbate can function as a donor intransplasma membrane electron

transport,

aswas 3Abbreviations: ferricyanide, K3[Fe(CN)6]; ADH, yeast alcohol dehydrogenase; BPDS,bathophenanthrolinedisulfonate; DCIP, 2,6-dichlorophenolindophenol; DCIP-sulfonate, 2,6-dichlorophenolin-dophenol-3'-sulfonate; PCMS,p-chloromercuriphenylsulfonicacid. 1178

TRANSPLASMA MEMBRANE ELECTRON TRANSPORT

reported to be the case with plasma membrane-enriched

fractions fromcottonandradish (15). MATERIALS AND METHODS PlantMaterial

Four-week-old sugar beet (Beta vulgaris L.) plants were

kindlysupplied by Hilleshog AB, Sweden and maintained in soil in a greenhouse with supplementary light (23 W m-2,

350-800 nm; Philips G/86/2 HPLR 400 W, The Nether-lands). Leaves of 6-to8-week-oldsugarbeetplantswereused.

The soil contained full nutritional requirements including iron. During Julyto September, sugarbeet leaveswere

har-vestedin the field.

Preparation of Plasma Membranes

Plasma membranes were purified from microsomal

frac-tions(10,000-50,000gpellet) by partitioning in an aqueous

polymer two-phase system as described (21). These plasma

membranes(>90% right-side-out vesicles)werealsoused for preparation ofinside-outplasmamembrane vesicles(21). Preparation of Loaded Plasma Membrane Vesicles

Lots of 12 gof leaves (midveins removed)were

homoge-nizedin 30mL of 330mm sucrose,50mmHepes-KOH (pH 7.5), 0.1% (w/v) BSA (Sigma A 3294, protease free), 2 mM

ascorbic acid, 1 mmDTT, 0.3% (w/v)insoluble PVP,20 Mm

leupeptin (SigmaL2884), and0.5 mmPMSF(addedin 150

ML

2-propanol;

omitted

during

preparation

of

plasma

mem-branes loaded with alcohol dehydrogenase plus NAD+) in a

small Sorvall blender kept on ice. Further additions to the homogenization medium werespecific foreach purpose:

so-dium ascorbate (final concentration, 100 mM)wascarefully

dissolved eitherbeforeorafterhomogenization for

prepara-tion ofascorbate-loadedand controlplasmamembrane vesi-cles, respectively. During preparation ofplasma membrane vesicles loaded with the NADH-generating system, the ho-mogenization mediumincluded 160 units mL-1 ADH (Boeh-ringer 102 717) and 10mMNAD+ (SigmaN7004).Controls

with only 10 mm NAD+ were also prepared. Loaded right-side-outplasmamembrane vesicleswerepurifiedfrom 10,000 to 50,000g microsomal fractions in the same way as

non-loadedplasmamembrane vesicles(21) by partitioningin an

aqueous polymer two-phase system (6.5% [w/w] Dextran T500,6.5% [w/w]polyethylene glycol 3350,330mmsucrose,

5 mMKCl, 5mm K-phosphate [pH 7.8]).Theplasma

mem-branefractionswerediluted with 330mmsucroseand5 mM

K-phosphate (pH 7.8),pelletedat100,000gfor1h,andfinally

resuspended inthe samemedium. DTT and other reducing substanceswerenotaddedduringthe latter part of the

prep-aration since they would interfere with the measurements (especially important when vesicleswere loaded with

ascor-bate). The yield ofplasma membranes was not changed by the loadingprocedures. About 90% of the NADH-ferricya-nide reductaseactivityofthese vesicleswaslatent (i.e.could

be measured only in the presence of Triton X-100), which shows that thesepreparationscontainedmainly sealed right-side-outvesicles(1, 21).

Reduction of External Electron Acceptors byLoaded Vesicles

Reductionof external electron acceptors was measured with anAminco DW 2 spectrophotometer operated either in the

split- or dual-wavelength mode (depending on the electron acceptorused, see below) at 22°C.Afterrecording a baseline, plasma membranes (0.1 mg protein or as indicated) were

addedtoastirredcuvette(Hellmacuv-o-stirmodel 333, FRG)

containing 330 mm sucrose, 25 mm Hepes-KOH (pH 7.5), and electron acceptor [sameconcentrations asduring meas-urement ofNADH-(acceptor) oxidoreductase activities (see

below)] in a final volume of2.5 mL, and the course of the

reduction recorded. Further additions were as indicated in

figuresand legends. For theextinctioncoefficients and wave-lengths usedfor the different acceptors, see below.

NADH-(Acceptor) ReductaseActivities

NADH-ferricyanidereductaseactivitywasmeasured

essen-tiallyaspreviously (1). Theassay was run at22TC in 1 mLof 330mm sucrose, 0.2 mM

K3[Fe(CN)6],

25 mm Hepes-KOH

(pH7.5), 0.25 mM NADH, 40

Mg

protein,and±0.025%

(w/

v) TritonX-100. The reaction wasinitiatedby theaddition

ofNADH.Correctionwasmadefor non-enzymatic reduction of ferricyanide. NADH-(acceptor) oxidoreductase activity

witheitherDCIP, DCIP-sulfonate, or ferric citrate as accep-torswasmeasured in thesameway,butwith 15

gM

DCIPor

DCIP-sulfonate, or250

gM

ferric citrate-Tris (prepared asin Buckhout etal. [9]) plus 50 Mm BPDS, respectively, instead offerricyanide. NADH-Cytcreductaseactivitywasmeasured

similarly

with 40 Mm Cytc

(Sigma

C 7752)asacceptor, and with 0.4

uM

antimycin A (SigmaA 2006) and 1 mm KCN presentinthe assaymedium. This activitywasdetermined± 0.015% (w/v) Triton X-100. The extinction coefficients and

wavelengths used for the different electron acceptors were:

DCIP-sulfonate, 25

mm-'

cm-' at645 nm; DCIP, 16

mM-cm-'

at600nm;Cytc, 19mM-'

cm-'

at550minus 600nm;

ferricyanide, 1 mM-'cm-'at420minus 500nm;Fe2+-BPDS,

22mM-'cm-' at535 nm. Protein

ProteinwasmeasuredaccordingtoBearden(6),withBSA asstandard.

RESULTS

Production of InternalNADHbyRight-Side-Out Plasma Membrane VesiclesLoaded with anNADH-Generating

System

Toinvestigateif electronscould betransferred from NADH acrosstheplasmamembranetomembrane-impermeable elec-tron acceptors at the apoplastic surface, NAD+ and ADH wereincluded in the homogenization medium and right-side-out plasma membrane vesicles subsequently prepared (Fig. 1). Weloaded the vesicles withanNADH-generatingsystem ratherthan with NADH forthe followingreasons: (a) The

formation ofNADHinside the vesiclescanbemeasured(see

below). Thus,we can be surethatNADH is presentduring

Acceptorox

NADH~~~

Q.~Acceptorred,

A/ H

E(OH

NADD

Figure 1. Schematic drawing of the NADH-generating system trapped inside right-side-out plasma membrane vesicles, and the reduction of an external acceptor via a presumptive transplasma membrane redox chain. The vesicles were loaded with 160 units mLV'yeast ADHand10mmNAD+ byinclusionof these substances duringhomogenization ofthe leaves. Ethanol(EtOH, 80mM, 0.5% [v/v]) added tothe cuvette will penetrate theplasmamembrane and lead to production of NADH inside thevesicles. Theoretically, the concentrationofNADHinsidethe vesicles would be0.4mm,

assum-ingthat ethanolconcentration andpHarethe same inside the vesicles asoutside,andusinganequilibriumconstantof 8.10-12M(5).

theexperiment, which is

particularly

important

ifresultsare negative, i.e. when the external electron acceptor isnot

re-duced.

(b)

The reaction can be started

(by

the addition of

ethanol)afterastable baseline has been established.

(c)

Since theconcentration ofNADH inside the vesicles is

kept

con-stant the rate of reduction of external electron acceptors

should belinear with time. The lattertwofactors also make itpossibletodeterminelow ratesaccurately.

The plasma membrane vesicles loaded with the

NADH-generating

systemproducedNADHwhenethanolwasadded: Addition of ethanol (80 mM; 0.5%

[v/v])

to these vesicles resulted in a transient increase in absorbance at 340 nm,

indicatinga production ofNADH

(Fig.

2A; AA ~ 7.

10-3).

Addition ofalargeramountofethanol gavealarger increase in absorbance (data not

shown),

but the concentration was usually kept below 0.6%

(v/v)

to ensure that membrane

integrity was maintained. No increase in absorbance was observed with vesicles loaded

only

with NAD'

(Fig. 2B).

External NAD'(1

mM)

was notreducedby the vesicles loaded with ADH plus NAD' (Fig. 2C, left) until Triton X-100 (0.025%

[w/v])

wasaddedtoreleasethe ADH

trapped

inside thevesicles

(Fig. 2C,

right).

This shows that

enclosed,

andnot

externally

bound,

ADHwas

responsible

for the ethanol-stim-ulatedNADH

production

with these vesicles

(Fig.

2A)

and

confirmsthatthis NADH was produced insidethe vesicles. Assumingthat theethanolconcentrationin thevesicles isthe same as the external concentration and that the pH in the

vesicles isbuffered, the concentration ofNADH inside the

vesicleswould be 0.4 mm under theprevailing conditions (see Fig. 1;calculatedusinganequilibriumconstantof8.

10-12

M

[5]). Thisismorethan 10timeshigher than the

Km(NADH)

fortheplasmamembrane

NADH-ferricyanide

reductase(1). Thus, thetrappedNADH-generatingsystem shouldbe able tosupporta transplasmamembrane NADH-oxidoreductase with electronstobetransferred to

ferricyanide

orother elec-tronacceptorsin theoutermedium

(Fig. 1).

Reduction ofExternalElectronAcceptors by

Right-Side-OutPlasmaMembrane Vesicles Loaded with the

NADH-GeneratingSystem

Addition of ethanol to right-side-out plasma membrane

vesiclesloaded with ADH and NADI led to a reduction of

external DCIP-sulfonate, suggesting a transfer of electrons across the plasma membrane from NADH to this acceptor

(Table I). Valinomycin (aK+ionophore) stronglystimulated

thisreaction whereasnigericin (an ionophoreexchanging K+ forHI)had noeffect(Fig. 3, A-C),indicating that the ethanol-stimulatedreduction of DCIP-sulfonate was an electrogenic process. This stimulation by valinomycin seems to exclude

thepossibility that the reduction ofDCIP-sulfonate wasdue to a slow permeation of this compound across the plasma membrane with aconcomitant reduction atthe cytoplasmic surface. No reduction ofexternalferricyanide, ferric citrate,

or Cyt c was observed, however (Table I), irrespective of whether or notvalinomycin ornigericinwas present(Fig.3D, datashown for ferriccitrate only). A very low rateofreduction

wouldhave beendifficulttodetectwith ferricyanidesince it has amuchlowerextinctioncoefficientthanDCIP-sulfonate.

Any reduction of ferric citrate would have been observed, however. Thus, neither ferricyanide nor ferriccitrateseemed tobe able to accept electrons from NADH via transplasma membrane electron transport.

Allacceptorstested were reduced at high rate by inside-out

plasma membranevesicles inthe presence ofNADH viaan electrontransport chain with both donor and acceptor sites onthecytoplasmicsurface oftheplasmamembrane(Table

I;

EIO EtOH EtOHTTX-100| / 00H1=n~lNAD+ / -0.006(AB)

V

A340-430

=0.03

(C)

-I 5 min

Figure2. Effect of ethanol (EtOH, 80 mm, 0.5% [v/v]) on NADI reduction withright-side-out plasma membrane vesicles (0.2mgof protein in 1 mL) loaded with(A)NADI and ADHaccording to Figure 1, and (B) NADI only. (C) Latency of the ADH activityin plasma membrane vesicles loaded according to Figure 1 as revealed by addition ofTritonX-100(0.025%[w/v])and NADI(1 mm). Seetext

TRANSPLASMA MEMBRANE ELECTRON TRANSPORT

Table 1. Reduction of Different Membrane-impermeable Electron Acceptors, and theMembrane-PermeableDCIP, with Right-Side-Out PlasmaMembrane Vesicles andInside-Out Plasma Membrane Vesicles

Right-side-out plasma membrane vesicles were loaded with an NADH-generating system (approximately 0.4 mm NADH inside the vesicles with 80 mm ethanol [Fig. 1]; 100 gg protein in 2.5 mL). Inside-out plasma membrane vesicles were in the presence of 0.25 mM NADH(40 Mg protein in 1 mL; no detergent present). Data are means ± SD obtained with two or more independent membrane preparations, except data obtained with DCIP (one membrane prep-aration only).

Activity

ElectronAcceptor NADH generated inside ExtemalNADH plus

right-side-outvesicles inside-out vesicles nmolacceptorreducedmin-'(mgprotein)'

Femcyanide 0 1030±280

DCIP-sulfonate 2.1 ±0.1 130±65

DCIP 34 120

Femccitrate 0 27±0

Cytc 0 200±58a

Adaptedfrom Askerlund et a/. (1).

1, 21). Membrane-permeable electron acceptors, such as

DCIP(14), werethereforereduced at a high rate also by

right-side-out vesicles loaded with the NADH-generating system (Table I). Triton X-100 completely inhibited this reaction

since it disrupted the vesicles leading to dilution of the

NADH/NAD'trappedinside the vesicles (Fig. 3E).

One reason notransplasma membrane electron transport to ferricyanide, ferric citrate, or Cyt c was observed could have been that the high concentration of NADI (10 mM) inside the vesicles completely inhibited the activity. With

inside-outplasma membranevesicles, however, 10mMNAD+

inhibited NADH-ferricyanidereductase to only about 50% in the presence of 0.4 mm NADH (data not shown). Thus,

assuming that thetransplasma membrane oxidoreductase is

similarly affected byNAD+,it would stillbeoperatingunder

theseconditions, althoughat a 50% reduced rate.

Incontrast to ourfindings,

Bcttger

(7)reportedareduction ofexternalferricyanide with right-side-out plasmamembrane vesicles from soybean hypocotyls loaded with NADH by

electroporation. This reaction was inhibited byactinomycin

D,whichalso inhibits the NADH oxidaseand NADH-ferri-cyanide reductase activities of these vesicles (18, 19). With

sugarbeetleaf plasmamembranevesicles,however, wefound

no effect of actinomycin D (15

AuM)

on NADH-(acceptor)

oxidoreductase activities with inside-out orsolubilized

vesi-cles,nordidweobserveanyeffectonDCIP-sulfonate reduc-tion with

right-side-out

vesiclesloadedwith the

NADH-gen-eratingsystem(datanotshown).Similarly,wefoundnoeffect ofp-nitrophenylacetate(0.1 mM),anotherinhibitor of plasma

PM 0.4% EtOH =0.01(A-D) AA =.0.064(E) 250s(A- D) 160s(E) PM Val. 02% in0.4% EtOH EtOH

Ferric citrate DCIP

Figure 3. Effectof ethanol(EtOH)onthe reduc-tion of(A-C)externalDCIP-sulfonate,(D) ferric citrate, and(E)DCIP with right-side-outplasma membrane vesiclesloaded with an NADH-gen-eratingsystem accordingtoFigure 1.The rate

recorded after addition of plasma membranes (PM)was set tozero,and all values are calcu-lated relativetothisbaseline. Indicatedrates are

nmol acceptor reducedmin-1(mgprotein)-'.The final concentrations ofvalinomycin(Val.)and

ni-gericin(Nig.)were40 and100 ngmL-', respec-tively. Theamountofplasma membrane added was: Ato C, 36 Mg protein; D, 73Mg; and E,

1004g.

D.

_

PM 0.4% Val. Nig EtOH inO.2% in0.2% EtOH EtOH t PM 0.3% EtOH TX-100 1181

A. control +TX-100 DCIP ElJ C. - ascorbate ascorbate+TX-100 D. ~ ~ ~ ~ ~ ~ ~ E. DCIP-sulfonate \ scorbate VV~I' ascorbate F Ferric c control G. Sitrate

Figure 4. Reduction of(A-C) DCIP, (D-E) DCIP-sulfonate, and (F, G) ferric citrate with ascorbate-loaded (100 mM) right-side-out plasma membrane vesicles(B, C, E, F) and control vesicles (A, D, G).Triton

X-100(0.025% [w/v])wasincludedasindicated. Arrows show where plasma membranes (0.1 mgprotein) wereadded tothe stirred

cu-vette. In Fand G theaddition of plasma membranes led toamuch larger increase in absorbance (light-scattering) than is evident from the figure (baselinereadjusted). The baseline wasnotreadjusted in Dand E.

membraneredox activityin soybean (18, 19). However, 0.1

mMPCMS (aSH-reagent) stronglyinhibited

NADH-depend-entredoxactivities with bothsugarbeet leaf(datanotshown) andsoybean hypocotyl (18, 19) plasmamembranes.Thus,we

observed 90% inhibition of DCIP-sulfonate reduction with vesicles loaded with the NADH-generating system, and a

similar inhibition of DCIP-sulfonate and NADH-ferric citrate reductase activities with both inside-out vesicles and Triton X-100-solubilized plasma membranes (data not shown). PCMS is supposedly membrane impermeable. The inhibition of DCIP-sulfonate reduction observed with the right-side-out vesicles shouldtherefore be duetomodification of SH-groups exposed on the outer surface of the plasma membrane.

Reductionof ExternalElectronAcceptors with Right-Side-Out Plasma Membrane Vesicles Loaded with Ascorbate

In search for other electron donors that might support transplasma membrane electron transport, right-side-out plasma membrane vesicles were prepared after inclusion of

100mm sodium ascorbate in the homogenization medium (controlvesicleswereprepared by includingthesameamount of ascorbateafter,rather than before homogenizationof the leaves).Theamountofascorbatetrappedinthevesicles could be measured by recording the reduction of the membrane-permeable electron acceptorDCIP (Fig. 4). Control vesicles reduced only trace amounts of DCIP (Fig. 4A), whereas

loaded vesicles reduced relatively large amounts (Fig. 4B), showing that the loading was successful. Triton X-100 (0.025% [w/v]) increasedthe rateof DCIP-reductionby about

100% withthe ascorbate-loadedvesicles(cf Fig.4, Band C)

showing that the penetration of DCIP through the plasma membrane was ratelimiting.Assumingthat theconcentration of ascorbate inthevesicleswasidenticaltotheconcentration

in thehomogenization medium(100 mM)and that all ascor-batewasfully oxidizedtodehydroascorbatewhenexposed to DCIP, an internal vesicle volume of 1.6

gL

(mg

protein)-'

canbecalculated.This is much less than the volumes ofsugar beetleaf plasma membrane vesiclesreported byLemoineand Delrot [4.7

gL

(mg

protein)-';

16] and Bush [9.6

/AL

(mg

protein)-';

10],andslightlyless than the volume reported by Buckhout [2

,uL

(mg

protein)-';

8].Thismay suggest that part

ofthe ascorbate was oxidizedduring the preparation proce-dure.However, asufficientamountofascorbateremained in the reduced state to make it possible to measure electron transport(Fig.4).

Thehydrophilicelectron acceptor DCIP-sulfonate was re-ducedbytheascorbate-loadedplasma membrane vesicles in

the absence of Triton X-100 (Fig. 4E, Table II), indicating

thatatransplasma membrane electrontransportsystem was

operating.

Therate wassimilartothatobtained with vesicles loadedwith the NADH-generatingsystem(cfTableI).Also,

ferriccitratewasreduced bythe ascorbate-loadedvesicles, but at about half the rate (Fig. 4F, Table II). To exclude the

possibility that the reduction of DCIP-sulfonate and ferric

citratewasdue toleakage ofascorbatefromthevesicles,

right-side-out vesicleswereloadedwith

['4C]ascorbate.

No leakage wasobserved 6 hafterresuspension of the vesicles (Fig. 5).

Valinomycin stimulated the reduction of DCIP-sulfonate

with the ascorbate-loaded vesicles by about 40% (data not

shown), and toa lesserextentthe reduction of ferric citrate

(about10%,datanotshown). Thus, much lessstimulationby

valinomycinwasfound inthese cases compared to reduction

of external DCIP-sulfonate by internalNADH(Fig. 3, A-C). Thereduction of ferric citrateandDCIP-sulfonate by the

ascorbate-loaded vesicles was not inhibited by PCMS (0.1

mM;

data notshown), incontrast tothe reductionof

DCIP-sulfonate by vesicles loadedwith the NADH-generating sys-tem which was 90% inhibited. There was no effect of

p-nitrophenyl acetic acidoractinomycin DonDCIP-sulfonate

reductionby the ascorbate-loadedvesicles (data not shown; notinvestigated with ferric citrate).

Table II. Reduction of External Ferric Citrate andDOIP-Sulfonate

withRight-Side-OutPlasmaMembrane Vesicles Loaded with Ascorbate(100mM)and withControl Vesicles(0.1 mg Protein with BothMaterials)

Initialrates are shown. Data are means±SD obtained with three independentmembranepreparations.

Activity ElectronAcceptor

Ascorbate-loaded vesicles Control vesicles nmolacceptorreducedmin-'(mgprotein)-' Ferric citrate 1.2±0.5 0.2±0.1 DCIP-sulfonate 2.0±0.7 0.2±0.1

T=0.1

(A-C)

A=0.008

(D-G) i 0 400s(A-C) 200s(D-G)

TRANSPLASMA MEMBRANE ELECTRON TRANSPORT 150 E U c N an S 100 . 50 -0 100 200 300 400

Time after resuspension, min

Figure 5. Experiment carried out to investigate if therewas any

leakage of ascorbate from the loaded plasma membrane vesicles. Twelvegramsof leaveswerehomogenized in homogenization buffer plus100mmascorbate and 0.82 MBqL-[carboxyl-14C]ascorbicacid (Amersham), and right-side-out plasma membrane vesicleswere sub-sequently preparedasdescribedin"Materials andMethods."Vesicles

werepelleted (10 min centrifugation inaBeckmanairfuge ofa0.16 mLsample containing 27 jigprotein)atdifferent times after

resus-pension (time 0), and the amount of radioactivityin the pellet and

supernatantweremeasuredbyscintillationcounting.The data indi-catethat ascorbate didnotescapefrom the loadedplasmamembrane

vesicles, unlessadetergentwasadded tothesuspension (far right; 0.025%[w/v] TritonX-100).

DISCUSSION

Transplasma membrane elecron transportwith NADH as

donorwas only detected when DCIP-sulfonate wasused as

acceptor(Table I; Fig. 3, A-C).In contrast toBottger (7),we were notable todetectanytransplasmamembrane electron transportfromNADH toferricyanide;neithertoferric citrate

nor to Cyt c (Table I, Fig. 3D). The reason only

DCIP-sulfonate was reduced could be its relative hydrophobicity. AlthoughDCIP-sulfonate as awhole isstrongly hydrophilic, the electron-accepting part of the molecule (the DCIP) is hydrophobic. Consequently, DCIP-sulfonate may be able to reachelectronsdeeperinto the membrane thanferricyanide, ferric citrate, orCyt c, of which the electron-accepting part

(the

Fe3")

ismore orless buried inahydrophilicshell. Results

obtained with intact plant tissues and cells, including leaf segments from sugarbeet, strongly suggestthattransplasma membrane electron transporttoacceptors suchasferricyanide

doesoccur(2, 9, 12, 17).Thepossibilityexiststhatascorbate rather than NADH is the electron donor in these cases,

although NADH isprobably the donor for thetransplasma membrane electrontransportsysteminiron deficienttomato roots (9). Alternatively, the NADH-system does not survive thepreparationofplasmamembrane vesicles. Forinstance,a

peripheral proteinontheapoplasticsurfacelinkingelectrons to more hydrophilic acceptors, such as ferricyanide, could easily be lost during membrane preparation. Giannini and Briskin (13), working witha crudeplasma membrane

frac-tion from red beet storage tissue, also found no evidence

for transmembrane electron transport from NADH to ferricyanide.

Using right-side-out vesiclesloaded with ascorbate we could detect transplasma membrane electron transport to both DCIP-sulfonate and ferric citrate (Table II; Fig. 4, E and F). Electron transfer from enclosed ascorbate to external ferricy-anide was earlier reported to occur in plasma membrane-enriched fractions from cotton roots and germinating radish seeds(15). These investigators foundthatferricyanide estab-lished a membrane potential (inside positive; measured by the accumulation ofthiocyanate) when added to ascorbate-loadedvesicles ofa presumedmixed orientation.

Of the inhibitors tested only PCMS showed any effect. Transplasma membrane electron transport from NADH to

DCIP-sulfonate was almost completely inhibited by PCMS, whereas transmembrane electron transport from ascorbate to

DCIP-sulfonatewas notaffectedatall. Provided that PCMS

really ismembrane-impermeable,this suggests that the accep-torsites on theapoplasticsurface differ for electron transport from NADH and ascorbate.This is supported by the fact that ferric citrate wasreducedbyascorbate(TableII,Fig.4F) but notby NADH (Table I,Fig. 3D) viatransmembrane electron transport.Possibly,the entire electron transportchainsdiffer,

sincedifferent donorsites should also berequired for ascor-bateand NADH. Inthe electrontransportchainusing ascor-bate as donor, the plasma membrane-bound b-Cyt with a

midpoint potential of150 mV(3)mayfunctionas anelectron

carrier, by analogy with the situation inchromaffingranules

(20).

Aspredicted from earlierresults (1, 4) transplasma mem-brane electron transport constitutesonlyasmall part of total

plasma membrane electrontransportactivity (atmostabout

10%;DCIP-sulfonate reduction byNADHinthe presenceof valinomycin

[cf.

Fig. 3, A-C and Table I]), the completely dominating activity occurringintheplaneof the membrane

withbothdonorand acceptorsites locatedonthecytoplasmic surface(1). The transplasmamembraneactivities recordedin the present work should, however, be sufficient to be of physiological significance.

ACKNOWLEDGMENTS

Wewishto thank Mrs.Ann-ChristineHolmstromand Mrs.Adine Karlssonforpreparation ofplasmamembranes,and Mr. K.Jonsson, NorraNobbelov,Sweden,forsupplyingsugarbeet leavesduring July toSeptember.We are alsogratefultoProfessorA.Trebst, Bochum, Germany,andProfessorG.Hauska, Regensburg,Germany, for their kindgifts of dichlorophenolindophenol sulfonate,toDr.T. J. Buck-hout,Kaiserslautern, Germany, forvaluablesuggestions,and to Dr. J.0.D.Coleman,Oxford, U.K., forcriticallyreadingthemanuscript.

LITERATURE CITED

1. AskerlundP,LarssonC,WidellS (1988)Localizationof donor and acceptor sitesofNADH dehydrogenase activitiesusing inside-outandright-side-out plasmamembranevesicles from plants. FEBS Lett 239: 23-28

2. AskerlundP,LarssonC(1989)Redoxactivitiesmeasured with inside-outandright-side-outplasma membranevesicles from sugar beet leaves. In J Dainty,MI DeMichelis, E Marr6, F Rasi-Caldogno,eds, Plant Membrane Transport: The Current Position.Elsevier, Amsterdam,pp 43-47

3. AskerlundP, Larsson C, Widell S(1989) Cytochromesof plant

supernatant

.0** . ,

in-

2

plasma membranes.Characterizationby absorbance difference spectrophotometryand redox titration. PhysiolPlant76:

123-134

4. AskerlundP,LaurentP,NakagawaH,Kader J-C(1991) NADH-Ferricyanidereductaseof leaf plasmamembranes.Partial pu-rification andimmunological relation topotatotuber micro-somalNADH-ferricyanidereductase andspinachleaf NADH-nitratereductase. PlantPhysiol 95:6-13

5. Barman TE(1969) EnzymeHandbook, Vol 1.Springer,Berlin, p23

6. BeardenJC Jr(1978)Quantitation ofsubmicrogramquantities ofprotein byanimprovedprotein-dye bindingassay. Biochim BiophysActa533: 525-529

7. Bottger M(1989)Transmembrane electrontransfer ofNADH loadedright sideoutvesicles. InJDainty,MIDeMichelis,E MarrE, FRasi-Caldogno,eds, PlantMembraneTransport:The CurrentPosition.Elsevier, Amsterdam,pp 55-60

8. Buckhout TJ(1989)Sucrose transport inisolatedplasma mem-branevesicles fromsugar beet(BetavulgarisL.).Evidencefor anelectrogenic sucrose-protonsymport. Planta178: 393-399 9. Buckhout TJ, Bell PF, LusterDG, ChaneyRL(1989)Iron-stress

induced redox activity in Tomato (Lycopersium esculentum Mill.) is localizedontheplasmamembrane. PlantPhysiol 90: 151-156

10. Bush DR (1989) Proton-coupled sucrose transport in plasma-lemmavesiclesisolated from sugarbeet(Beta vulgarisL. cv GreatWestern)leaves. PlantPhysiol89: 1318-1323

11. Crane FL,LowH,Clark MG(1985) Plasma membrane redox enzymes. In ANMartonosi, ed, The Enzymes of Biological Membranes.Plenum Press, NewYork,pp465-510

12. DharmawardhaneS, SternAI, Rubinstein B(1987) Light-stim-ulatedtransplasmalemmaelectron transport in oatmesophyll cells. Plant Sci51: 193-201

13. GianniniJL, Briskin DP(1988) Pyridine nucleotide oxidation

byaplasmamembranefractionfrom red beet(Beta Vulgaris L.)storagetissue.ArchBiochemBiophys260:653-660 14. HauskaG,TrebstA,Draber W(1973) Lipophilicityandcatalysis

ofphotophosphorylation. II.Quinoid compoundsasartificial carriers in cyclic photophosphorylation and photoreductions

by photosystemI.BiochimBiophysActa 305:632-641 15. Hassidim M, Rubinstein B, Lerner HR, Reinhold L (1987)

Generationofamembranepotential byelectron transport in plasmalemma-enrichedvesiclesfromcottonandradish.Plant Physiol85:872-875

16. LemoineR,DelrotS(1989)Proton-motive-force-drivensucrose

uptake in sugar beet plasmamembrane vesicles. FEBS Lett 249: 129-133

17. M0llerIM,Crane FL(1990)Redox processes inthe Plant Plasma Membrane. InCLarsson,IM M0ller, eds,The PlantPlasma Membrane.Structure,Function andMolecularBiology. Sprin-ger,Berlin,pp93-126

18. MorreDJ, Brightman AO,WuL-Y,BarrR,LeakB,CraneFL (1988)Roleofplasmamembrane redoxactivitiesinelongation growthinplants.Physiol Plant 73: 187-193

19. MorreDJ,CraneFL,BarrR,PenelC,Wu L-Y(1988)Inhibition ofplasma membrane redoxactivities and elongationgrowth ofsoybean. PhysiolPlant 72:236-240

20. Njus D, KnothJ, Cook C, Kelley P (1983) Electron transfer

acrossthechromaffingranulemembrane. J Biol Chem 258: 27-30

21. PalmgrenMG,AskerlundP,Fredrikson K, WidellS,Sommarin M, Larsson C (1990) Sealed inside-out and right-side-out plasmamembrane vesicles. Optimalconditions forformation andseparation.PlantPhysiol92:881-890

22. Palmgren MG, SommarinM, UlvskovP,Larsson C (1990)Effect ofdetergents on the H+-ATPase activity ofinside-out and right-side-out plant plasma membrane vesicles. Biochim Bio-physActa1021:133-140