Proc. Natl. Acad. Sci. USAVol. 73, No. 7, pp. 2211-2215, July 1976Biochemistry

Redox titration of fluorescence yield of photosystem II(photosynthesis/primary electron acceptor/photophosphorylation)

BACON KE, FRED. M. HAWKRIDGE*, AND SAURA SAHUtCharles F. Kettering Research Laboratory, Yellow Springs, Ohio 45387

Communicated by Martin D. Kamen, April 5, 1976

ABSTRACT The variable fluorescence yield of photosystemII is dependent on the redox state of the fluorescence quenchermolecule or the primary electron acceptor of the system. Wehave carried out redox titrations of fluorescence yield of aphotochemically active photosystem-II reaction-center particleand have measured the redox potential of the photosystem-Ilprimary acceptor.During reductive titrations using dithionite as the reductant,

only a single quenching transition was observed. For instance,at pH 7.0, the midpoint potential of the fluorescence transitionis -325 mV, and those at a pH between 6.0 and 7.5 are consistentwith a pH dependence of about 60 mV/pH unit. At a given pH,the midpoint potential of the transition closely corresponds tothat of the most negative transition reviously measured inunfractionated chloroplasts (both by chemical reductive titra-tion). Oxidative titrations using ferricyanide as the oxidantyielded hysteresis in the titration curves.

Similar changes in fluorescence yield were observed in redoxtitrations by electrochemical reduction or oxidation. Electro-chemical reductive and oxidative titrations yielded reversibletransitions, contrary to the hysteresis observed during chemicaloxidative titration. From coulometric-titration data, we haveestimated that most likely one electron is involved in the redoxtransition of the fluorescence-quencher or primary-electron-acceptor molecule of photosystem IL These findings are con-sistent with the current proposal that a membrane-bound plas-toquinone functions as the primary acceptor of photosystemIL.

The photosynthetic apparatus in green plants consists of twophotosystems. Photon capture by each photosystem results inan energetically excited state leading to the formation of twooppositely charged species. The primary electron donors in bothphotosystems are probably made up of chlorophyll a moleculespresent in a specialized environment. Whereas the primarydonor of photosystem I, P-700, has been known for some time(1), the primary donor of photosystem II, P-690 (2, 3), and theprimary electron acceptor of photosystem I, P-430 (4), havebeen discovered only more recently.The primary electron acceptor of photosystem II, because

of its role in controlling the variable fluorescence of photosystemII, was first designated by Duysens and Sweers (5) as "Q," whichstands for the quencher of fluorescence. The quencher moleculeor the primary acceptor of photosystem II has often beenthought to be a plastoquinone (6), as exogenous quinones havebeen found to quench strongly the fluorescence of chlorophyllin chloroplasts and algae (7). The primary-acceptor plasto-quinone, however, must possess some unique properties dif-ferent from plastoquinone molecules in the secondary-acceptorpool (3).

Stiehl and Witt (8) observed a spectral species, "X-320",whose kinetic behavior was said to be consistent with that of thephotosystem-II primary acceptor. Knaff and Amon (9) observed

* Visiting Scientist (summer, 1975) from the Chemistry Department,University of Southern Mississippi, Hattiesburg, Miss. 39401.

t Present address: Department of Biochemistry, Duke University,Durham, N.C. 27710.

2211

another spectral species, "C-550", which also appeared to haveproperties of the photosystem-II primary acceptor (10). Morerecently, van Gorkom (11) and Pulles et al. (12) identifiedplastoquinone as the primary acceptor of photosystem II. Bothgroups found that the light-induced absorption changes asso-ciated with the photosystem-II primary acceptor apparentlyalso included absorption changes that have previously beenattributed to both X-320 (8) and C-550 (9, 10).The redox potential of the photosystem-TI primary electron

acceptor was first measured by Kok et al. (13), who monitoredthe fluorescence-yield changes as a function of the redox po-tential, and they reported a midpoint potential of +180 mV.Similar fluorescence titration of spinach chloroplasts by Cramerand Butler (14) showed quenching occurred at -35 mV and-270 mV (at pH 7), and the midpoint potentials were pH-dependent. The titration data could be interpreted as to rep-resent either two quencher molecules, each with a differentredox potential, or a single quencher molecule that undergoesa two-stage reduction. The authors favored the less negativepotential as the representative value for the primary electronacceptor of photosystem II, but the significance of the lowerpotential quencher remained undecided. Similar redox po-tential values of the photosystem-II primary acceptor have alsobeen estimated by titrating the absorption changes of C-550(10, 15). From an indirect estimate of the pK value of thephotosystem-II primary acceptor, Knaff (16) recently suggestedthat the acceptor may function with an effective potential of-130 mV.We report here data on the redox titration of fluorescence

yield from a photochemically active photosystem-II reaction-center particle isolated from spinach chloroplasts by Tritonfractionation (17-19). We have found that in all cases examinedonly one quenching transition occurred and, at a given pH, themidpoint potential of the transition closely corresponds to thatof the most negative transition previously found in unfrac-tionated chloroplasts (14). From electrochemical-titration datawe have estimated that most likely one electron is involved inthe redox transition of the quencher molecule.

EXPERIMENTAL

The photosystem-Il reaction-center particles were preparedfrom spinach chloroplasts by Triton fractionation using themethod described previously (17, 18). Chloroplasts used forduplicating the fluorescence titrations were prepared in thesame way as described in ref. 14. All chemicals were of reagentgrade when available and were used without further purifi-cation.

Chemical titrations using dithionite (Fluka, purum grade)as the reductant and ferricyanide as the oxidant were carriedout in a special 1 cm X 1 cm quartz cell described in detail re-cently (20, 21). The redox mediator concentrations are givenin the figure legends.

Dow

nloa

ded

by g

uest

on

Aug

ust 3

1, 2

021

Proc. Natl. Acad. Sci. USA 73 (1976)

Electrochemical titrations were carried out in the electro-chemical cell of Hawkridge and Kuwana (22). The cell wasmodified by replacing the tin-oxide window with a quartz plate,and a gold-foil electrode was fixed along the side wall of the cell.This cell has a path length of 12 mm and a total volume of 1.75ml. The chloroplast sample containing the redox mediators wasdeoxygenated in a separate reservoir bulb attached to the cellbefore it was transferred into the cell. A potentiostat assembledfrom instrument modules [McKee Pedersen Instruments (MPI),Danville, Calif.] or a PAR model 173 potentiostat (PrincetonApplied Research, Princeton, N.J.) was used for controlling thereduction or oxidation.The fluorescence was measured with a weak monitoring

beam [650 nm, 8-10 ergs/cm2-sec (8-10 mW/M2), isolated bya 10-nm wide interference filter (Baird Atomic), and modulatedby an elasto-optic modulator at 25 kHz (23)]. The modulatedfluorescence passing through Corning filters 2-64 and 4-77(cut-on at 687 nm) was detected by an EMI 9558 photomulti-plier tube and processed through a PAR model 211 preamplifierand a PAR model 220 Lock-In amplifier and recorded on achart recorder. When the 1 cm X 1 cm quartz cell was used inthe chemical titrations, the fluorescence emitted at right angleto the measuring beam was detected. When the electrochemicalcell (22) was used, the fluorescence emitted in line with themeasuring beam was detected from the backside of the cell.

RESULTSPotentiometric titration of fluorescence yield using achemical reductant or oxidantFluorescence induction in the photosystem-II particles has beenexamined previously (19) and correlated with their photo-chemical activities as a further confirmation of their photo-system-II character. The internal oxidant pool of photosystemII was titrated by light quanta and estimated from the fluo-rescence-rise curves (19). The redox properties such as the ox-idation-reduction potential can be determined by a potentio-metric titration of the fluorescence yield (13, 14). The inferencederived from both light and chemical titrations is based on theinterpretation of Duysens and Sweers (5), namely, the fluo-rescence yield is dependent on the redox state of the primaryelectron acceptor or the fluorescence quencher of photosystemHI.The photosystem-IT particle may contain only components

immediately associated with the reaction center. A potentio-metric titration could therefore yield a change less complicatedthan in chloroplasts, allowing more insight concerning the redoxcomponents in photosystem II. Fig. 1 shows the results of apotentiometric titration of the photosystem-IT particles, usinga monitoring beam at about 10 ergs/cm2-sec (10 mW/m2). Thetitration was carried out at four pH values from 6.0 to 7.5, athalf a pH-unit intervals, and from +100 to -400 mV (all po-tentials expressed relative to the normal hydrogen electrodeunless otherwise indicated); the negative potential limit wasrestricted by the prevailing pH of the medium. The magnitudeof fluorescence yield appears to depend on pH. This behavior,although its nature is unknown, parallels the pH dependencyof the photochemical activity of these particles (24).

In the reductive titration (from right to left in the figure), asingle, relatively simple change occurred at all pH values. Forinstance, at pH 7.0, the midpoint potential is -325 mV, andthose at other pH valubs are consistent with a pH dependencyof about 60 mV/pH unit. However, the slope of the titrationcurves suggests that the reaction may involve more than twoelectrons. The oxidative titration represented by the dashed

X! A

m I \\

._ \CV

a) I I I-I.-A

a) \

LI)

C

C.

a)

.0oLI)

-400 -200Eh, mv

0 +100

FIG. 1. Potentiometric titration of the fluorescence yield ofphotosystem-IT particles. See refs. 20 and 21 for detailed procedures.All samples contained about 10 gg of chlorophyll per ml in 0.1 Mphosphate buffer. The redox mediators were: 1,4-naphthoquinone(30 AM); 5-hydroxy-1,4-naphthoquinone (30 MM); 2-hydroxy-1,4-naphthoquinone (30MAM); anthraquinone-,8-sulfonate (0.1 mM); andneutral red (30 MM). Reductive (-) and oxidative --- -) titrationswere made with 0.1 M dithionite and ferricyanide, respectively, as thetitrant.

curves going from left to right in the figure are much less welldefined either in the pH dependency of the midpoint potentialsor in the slopes of the curves.The courses of titration observed previously for chloroplasts

are quite different from those observed here: for instance, atpH 7, two major breaks were observed during the reductivetitration, one at a much more positive potential of -20 mV, andthe other at -320mV (14). Similarly, a hysteresis was observedduring the oxidative titrations, i.e., the titration curves wereshifted toward a more positive potential (14) similar to thatshown in Fig. 1, but the breaks were reasonably well defined.

In view of the above-mentioned difference, we carried outa potentiometric titration of the fluorescence yield with un-fractionated chloroplasts, and the results are shown in Fig. 2.At pH 7.0, a reductive titration produced three transitions, atmidpoint potentials of +10, -235, and -321 mV. The firsttransition was always the largest at all three pH values exam-ined, and its slope appeared to correspond to a one-electronchange. The second transition was relatively small. The thirdtransition, occurring at the most negative potential, was of in-termediate size and its slope would suggest a redox transitioninvolving more than two electrons at all three pH values ex-amined. The transitions show a shift of 60 mV/pH unit. Hys-teresis also occurred during the oxidative titrations (cf. ref. 14).Interestingly, the most negative transitions observed during thereductive titration of chloroplasts at all three pH values corre-sponded closely with the single transitions observed in thephotosystem-II particles, both in the midpoint potentials as wellas the slope of the transitions.

2212 Biochemistry: Ke et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 3

1, 2

021

Proc. Natl. Acad. Sci. USA 73 (1976) 2213

a}(ii. _

m7a)'I-

a)0CDa)0

0

-400 -200 0 o100

Eh, mvFIG. 2. Potentiometric titration of unfractionated spinach

chloroplasts. All other conditions were identical to those for Fig. 1.

Potentiometric titration of fluorescence yield byelectrochemical reduction or oxidationWe further used electrochemical titrations to examine thefluorescence yield changes. The electrochemical behavior ofthe mediators in the presence of chloroplasts or subchloroplastparticles was first examined, prior to fluorescence yield mea-surements by recording a cyclic voltammogram of the samplesystem in the same electrochemical cell with which fluorescenceyield was to be measured. Fig. 3 shows typical cyclic voltam-mograms for samples containing unfractionated chloroplasts(at pH 7.0) and photosystem-Il particles (at pH 7.5), with theappropriate redox mediators present. The cyclic voltammo-grams show the expected reversible reductions and oxidations

0-

@,1 25u0-

o -.2 -.4 -.6 -.8 -1.O V(vs. Ag/AgCI)

FIG. 3. Cyclic voltammograms of chloroplast (top traces, at pH7.5) and photosystem-IT subchloroplast (bottom traces, at pH 7.0)suspensions containing the following: NaCl (0.2 M); indigotetrasul-fonate (10 IAM); 2-hydroxy-1,4-naphthoquinone (20 MAM); anthra-quinone-fl-sulfonate (4Q MM); neutral red (30 MM); 1,1'-trimethy-lene-2,2'-dipyridylium dibromide (0.5 mM); chloroplast (16 gg ofchlorophyll per ml); photosystem-II subchloroplasts (15 jig of chlo-rophyll per ml) in 0.1 M phosphate buffer.

b e

U_ ~~~~~~~~~~dTime or Accumulated Charge

FIG. 4. Fluorescence-yield changes in the photosystem-LI sub-chloroplast particles upon charge injection (a-b) and charge with-drawal (b-c). Trace d-e shows reduction during the fourth cycle.Abscissa scale is linear with respect to injected charge (see Fig. 5) butnonlinear with time; the constant fluorescence portion took ap-proximately 15 min, the rising fluorescence portion approximately5 min. Sample composition, same as in legend of Fig. 3. See text fordiscussion of other details.

of the mediators occurred in the presence of (sub)chloroplastswhen the voltage was scanned in the negative and positive di-rections, respectively.

After the cyclic voltammetric check, the potential of thegold-foil electrode was stepped to -900 mV [versus the Ag/AgC1 (1.0M KCI) reference electrode] and held constant at thatpotential by the potentiostat; the fluorescence yield was re-corded as a function of the accumulated charge injected intothe sample system. A' typical recording is shown in Fig. 4 forthe fluorescence yield change of the photosystem-II particlesat pH 7.0. In our present experimental arrangement, the actualpotential of the system at any given time was measured by anadditional platinum electrode placed in the cell. When thepotential of the platinum electrode was measured versus thereference electrode, all other leads were momentarily discon-nected from the potentiostat.

In the titration'shown in Fig. 4, the fluorescence yield of thephotosystem-II particle remained at a low level up to -540 mV(versus Ag/AgCl), then increased steadily and became saturatednear -590 mV. A slight drop in fluorescence yield was apparenteven when the potential was maintained constant. However,when the applied potential was stepped back to 0 mV (versusAg/AgCl) to oxidize the system, a sharp drop in fluorescenceyield occurred. The fluorescence-yield increase during the firstcycle occurred over a potential span of approximately 50 mV,while the fluorescence-yield decrease and subsequent fluo-rescence increases and decreases all occurred with an evensteeper slope (see Fig. 4). This was probably caused by a traceof oxygen present in the sample initially.The midpoint potentials estimated from the point-by-point

measurements during electrochemical titrations were consis-tently more negative than those obtained from the continuouslyrecorded values of the chemical titrations. The reason for thisdifference is not known. However, steep slopes in the fluores-cence transitions were obtained both by chemical (Fig. 1) andby electrochemical (Fig. 4) titrations, and the pH dependenceof the midpoint potentials appears to be the same, i.e., about60 mV per pH unit. On the other hand, the reverse titration byelectrochemical oxidation was markedly different from thoseobserved in the chemical titrations. Whereas the chemical ti-trations always showed hysteresis during the reverse oxidativereaction, as shown in Figs. 1 and' 2, and also reported previouslyby Cramer and Butler (14), electrochemical oxidative titrationsappear to be reversible and free of such hysteresis at all pHvalues we have examin'ed. It is possible that the oxidative titrant,ferricyanide, being a well-known electron acceptor for pho-

Biochemistry: Ke et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 3

1, 2

021

Proc. Natl. Acad. Sci. USA 73 (1976)

2,

a)

Voltage, 0.1 volt

FIG. 5. Coulometric titration of fluorescence yield of the photo-system-II subchloroplast particles in the modified electrochemicalcell (22). Sample composition same as in legend of Fig. 3. See text fordetailed discussion.

tosystem II, may not be able to effect a controlled oxidation ofthe fragments through the mediators.

Coulometric titration of fluorescence yieldBecause both the chemical and electrochemical titrationsyielded the unusual slope, which suggests that the redox tran-sition of the quencher or primary-acceptor molecule involvesmore than two electrons, we attempted to estimate the numberof electron changes more directly by coulometric titrations inthe modified electrochemical cell (22).

Since the course of the fluorescence rise is quite well definedand the onset of the fluorescence rise is sharp, the coulometricmeasurement was made by stepping the applied potential to-900 mV (versus Ag/AgCl). Immediately before the onset offluorescence rise, simultaneous recordings of the charge injectedand the electrolysis time were begun and the actual potentialsof the system at the beginning and end of the fluorescence risewere noted. Typical results of such a titration of the photosys-tem-II particle at pH 7.5 are shown in Fig. 5.The charge and time measurements were started at the po-

tential of -446 mV (versus Ag/AgCl) and ended at -605 mV.The total charge was determined by passing the electrolysiscurrent flowing through the working electrode first througha current-to-voltage converter and then integrating this voltagewith respect to time with a chopper-stabilized operationalamplifier (MPI 1031 unit). The total charge consumption ac-

companying the fluorescence rise was estimated from the dis-tance between the two intercepts of the curve with the ex-

trapolations of the initial and final portions of the curve (dashedlines). During the electrolysis, nonfaradaic events also con-

suming charges must be accounted for and corrected from thetotal charge. The nonfaradaic current was separately estimatedby running the same experiment in the absence of the redoxmediators. The net charge was estimated to be 0.64 micro-equivalent from three reductive titrations and 0.60 micro-equivalent from three oxidative titrations. The sample con-

taining the photosystem-IT particle had a chlorophyll concen-

tration of 17 ,uM. As the ratio of the concentration of the reac-

tion-center components (C-550 and P-680, as well as cyto-chrome bss9) to the total chlorophyll in this sample was pre-viously estimated to be Y4O (19), the primary-acceptor concen-tration was assumed to be the same, which would be 0.42 AuM.The number of electron equivalent per molecule is then 1.4-1.5.Comparable results were also obtained from a coulometric ti-tration of the photosystem-II particle at pH 6.5 in a thin-pathlength cell designed by Heineman and coworkers (25). It shouldbe noted that the photosystem-II fragments, though highlypurified, most likely also contain other components that mayalso undergo redox changes in the fluorescence transition re-

gion. Thus, the net electron charge measured here may not beaccounted for exclusively by the fluorescence change. With thisconsideration, we would estimate that the redox change of thequencher molecule most likely involves one electron.

DISCUSSIONGiven the fact that only a single redox transition was apparentin the fluorescence-yield change in the photosystem-II reac-tion-center particles at any given pH, and that the transitionagrees well with that observed at the most negative potentialin unfractionated chloroplasts, we are led to conclude that themore negative transition is representative of the quencher orprimary-electron-acceptor molecule of photosystem II. Themore positive transitions observed in chloroplasts probablyrepresent the secondary electron acceptors that are in a redoxequilibrium with the primary acceptor in the unfractionatedchloroplasts. The photosystem-II subchloroplasts presumablyare largely devoid of secondary acceptors (18, 19).The general view at present is that (bound) plastoquinone

is functioning as the primary electron acceptor of photosystemII (11, 12). Adopting this view and accepting the basic view thatthe fluorescence-yield change is a valid indication of the redoxstate of the photosystem-IT primary acceptor (5), despite nu-merous recent findings that the photosystem-II fluorescenceyield can also be quenched by a number of other components(3), we will now examine some implications of our presentfindings.

Although it is generally accepted at present that light- in-duced charge separation in photosystem II results in a verystrong oxidant and a relatively weak reductant, the red photonabsorbed by photosystem II, however, has sufficient energy toproduce a more negative reductant than currently envisioned.In protic solvents, one-electron reduction of quinones is noteasily observed, because the semiquinone radicals readily dis-proportionate into a fully reduced hydroquinone and a fullyoxidized quinone. The redox potential of such quinone reactionshas been reported to be near 100 mV (26). However, in aproticsolvents, semiquinones are very stable species. Furthermore,the redox potential of the one-electron reduction of quinonesin aprotic solvents is at a much more negative value, rangingfrom -300 to -500 mV (27). Thus, the relatively negative redoxpotentials observed for the fluorescence-yield increase attrib-uted to the quencher or primary-acceptor molecule of photo-system II is consistent with the notion that the plastoquinonemolecules, which are presumably tightly bound to the pho-tosynthetic membrane, are located in a hydrophobic environ-ment.

As the intersystem potential span, i.e., the potential differencebetween the oxidized P-700 and the reduced primary acceptorof photosystem II, is vital for photosynthetic ATP formation,the more negative redox potential observed for the system-IIquencher molecule may lead to a reassessment of the energeticaspect of photophosphorylation (2, 16).

Stiehl and Witt (28) found that plastoquinone in the secon-dary-acceptor pool was reduced to the hydroquinone form eventhough supposedly only one electron was transferred throughthe electron-transport chain when single turnover short flasheswere used for excitation. This anomaly was explained by as-suming that the system-II primary acceptor (designated as X)is present as a "twin," i.e., X-X, and each X is complexed witha primary-donor molecule (say, P-690). The photochemicalreaction leads to a doubly charged twin, X--X-, which reducestwo molecules in the plastoquinone (PQ) pool, and twoplasto-semiquinone molecules in turn disproportionate to onePQ and PQ2-. The half-life of the primary acceptor X was

2214 Biochemistry: Ke et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 3

1, 2

021

Proc. Natl. Acad. Sci. USA 73 (1976) 2215

found to correspond to the risetime of plastoquinone reduction(27).Our observation that reduction of the system-TI quencher

molecule involves one electron appears to be consistent withthe "twin" quencher scheme. Although each quencher mole-cule undergoes a one-electron reduction, the close associationof two such quencher molecules in a twin unit could lead tocomplex reaction kinetics and conceivably account for theunusual slopes of the fluorescence-yield titration curves.

We thank Dr. W. R. Heineman for many useful discussions re-

garding the use of the thin-layer electrochemical cell and Dr. P.Mohanty for helpful comments on the paper. We also thank Dr. Mi-chael J. Brown of the Imperial Chemicals Industries for a gift sampleof the 1,1'-trimethylene-2,2'-dipyridyfium dibromide and Mr. E. R.Shaw for providing the photosystem-II particle. This work was sup-

ported by a National Science Foundation Grant GB-29161. This iscontribution no. 550 from Charles F. Kettering Research Laboratory.

1. Kok, B. (1956) Biochim. Biophys. Acta 22,399-401.2. Trebst, A. (1974) Annu. Rev. Plant Physiol. 25,423-458.3. Radamer, R. & Kok, B. (1975) Annu. Rev. Biochem. 44,409-433.4. Ke, B. (1973) Biochim. Biophys. Acta 301, 1-3.5. Duysens, L. N. M. & Sweers, H. E. (1963) in Studies on Mi-

croalgae and Photosynthetic Bacteria, ed. by Japan Soc. PlantPhysiol. (Univ. Tokyo Press, Tokyo), pp. 353-372.

6. Amesz, J. (1973) Biochim. Biophys. Acta 301,35-51.7. Amesz, J. & Fork, D. C. (1967) Biochim. Blophys. Acta 243,

97-107.8. Stiehl, H. H. & Witt, H. T. (1969) Z. Naturforsch. Teil B 231,

220-224.9. Kn4ff, D. B. & Arnon, D. I. (1969) Proc. Natl. Acad. Sci. USA,

63,963-969.

10. Erixon, L. & Butler, W. L. (1971) Biochim. Biophys. Acta 234,381-389.

11. van Gorkom, H. (1974) Biochim. Biophys. Acta 347,439-442.12. Pulles, M. P. J., Kerhof, P. L. M. & Amesz, J. (1974) FEBS Lett.

47, 143-145.13. Kok, B., Malkin, S., Owens, 0. & Forbush, B. (1966) Brookhaven

Symp. Biol. 19,446-459.14. Cramer, W. A. & Butler, W. L. (1969) Biochim. Biophys. Acta

172,503-510.15. Knaff, D. B. (1975) Biochim. Biophys. Acta 376,583-587.16. Knaff, D. B. (1975) FEBS Lett. 60,331-5.17. Vernon, L. P., Klein, S., White, F. G., Shaw, E. R. & Mayne, B.

C. (1971) in Proc. H. Intern. Congr. Photosynthesis, Stresa (Dr.Junk N. V., The Hague), pp. 801-812.

18. Ke, B., Vernon, L. P. & Chaney, T. (1972) Biochim. Biophys. Acta256,345-357.

19. Ke, B., Sahu, S., Shaw, E. R. & Beinert, H. (1974) Biochim. Bio-phys. Acta, 347,36-48.

20. Ke, B. (1975) Bioelectrochem. Bioenerg. 2,93-105.21. Ke, B., Hansen, R. E. & Beinert, H. (1973) Proc. Natl. Acad. Sc.

USA 70,2940-2945.22. Hawkridge, F. M. & Kuwana, T. (1973) Anal. Chem. 45,

1021-1027.23. Breeze, R. H. & Ke, B. (1972) Anal. Biochem. 50,280-313.24. Vernon, L. P. & Shaw, E. R. (1969) Plant Physiol. 44,1645-1649.25. Norris, B. J., Meckstroth, M. M. & Heineman, W. R. (1976) Anal.

Chem., 48,630-632.26. Morton, R. A. (1965) in Biochemistry of Quinones, ed. Morton,

R. A. (Academic Press, New York), pp. 1-21.27. Mann, C. L. & Barnes, K. K. (1970) in Electrochemical Reactions

in Non-aqueous Systems (Marcel Dekker, Inc., New York), pp.190-197.

28. Stiehl, H. H. & Witt, H. T. (1969) Z. Naturforsch. Teil B 24,1588-1598.

Biochemistry: Ke et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 3

1, 2

021