Quantitation of the rapid electron donors to P700, the functional plastoquinone pool, and the ratio of the photosystems in spinach chloroplasts.

Recent studies of chloroplast architecture have emphasized the segregation of photosystem I and photosystem II in different regions of the lamellar membrane. The apparent localization of photosystem II reaction centers in regions of membrane appression and of photosystem I reaction centers in regions exposed to the chloroplast stroma has focused attention on the intervening electron carriers, carriers which must be present to catalyze electron transfer between such spatially separated reaction sites. Information regarding the stoichiometries of these intermediate carriers is essential to an understanding of the processes that work together to establish the mechanism and to determine the rate of the overall process. We have reinvestigated the numbers of photosystem I and photosystem II reaction centers, the numbers of intervening cytochrome b6/f complexes, and the numbers of molecules of the relatively mobile electron carriers plastoquinone and plastocyanin that are actively involved in electron transfer. Our investigations were based on a new experimental technique made possible by the use of a modified indophenol dye, methyl purple, the reduction of which provides a particularly sensitive and accurate measure of electron transfer. Using this dye, which accepts electrons exclusively from photosystem I, it was possible to drain electrons from each of the carriers. Thus, by manipulation of the redox condition of the various carriers and through the use of specific inhibitors we could measure the electron storage capacity of each carrier in turn. We conclude that the ratio of photosystem I reaction centers to cytochrome b6/f complexes to photosystem II reaction centers is very nearly 1:1:1. The pool of rapid donors of electrons to P700 includes not only the 2 reducing equivalents stored in the cytochrome b6/f complex but also those stored in slightly more than 2 molecules of plastocyanin per P700. More slowly available are the electrons from about 6 plastoquinol molecules per P700.

Recent studies of chloroplast architecture have emphasized the segregation of photosystem I and photosystem I1 in different regions of the lamellar membrane. The apparent localization of photosystem 1 1 reaction centers in regions of membrane appression and of photosystem I reaction centers in regions exposed to the chloroplast stroma has focused attention on the intervening electron carriers, carriers which must be present to catalyze electron transfer between such spatially separated reaction sites. Information regarding the stoichiometries of these intermediate carriers is essential to an understanding of the processes that work together to establish the mechanism and to determine the rate of the overall process. We have reinvestigated the numbers of photosystem I and photosystem I1 reaction centers, the numbers of intervening cytochrome bdf complexes, and the numbers of molecules of the relatively mobile electron carriers plastoquinone and plastocyanin that are actively involved in electron transfer. Our investigations were based on a new experimental technique made possible by the use of a modified indophenol dye, methyl purple, the reduction of which provides a particularly sensitive and accurate measure of electron transfer. Using this dye, which accepts electrons exclusively from photosystem I, it was possible to drain electrons from each of the carriers. Thus, by manipulation of the redox condition of the various carriers and through the use of specific inhibitors w e could measure the electron storage capacity of each carrier in turn. We conclude that the ratio of photosystem I reaction centers to cytochrome bdf complexes to photosystem I1 reaction centers is very nearly 1: 1: 1. The pool of rapid donors of electrons to P 7 M includes not only the 2 reducing equivalents stored in the cytochrome b,,Jf complex but also those stored in slightly more than 2 molecules of plastocyanin per P 7 0 0 . More slowly available are the electrons from about 6 plastoquinol molecules per PTo0.
A minimal description of photosynthetic electron transfer in eucaryotes must accommodate kinetic features of the individual electron transfer reactions as well as the structural features of orientation and distribution of electron transfer components in the lamellar membrane. The sequence of reactions in electron transfer, insofar as it is known, has been * This work was supported in part by National Science Foundation Grant PCM-8314461. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. established for many years with only relatively few recent embellishments (1). However, our knowledge of the sequential order of the charge transfer reactions may give a falsely optimistic view of our understanding of the mechanism of photosynthetic electron transfer if our information of stoichiometries and kinetic factors remains inadequate.
Recent progress in elucidating lamellar membrane architecture has appropriately focused attention on communication of the PS I' and PS 11 reaction center complexes with the intervening cytochrome be/[ complex. There are many independent lines of evidence which demonstrate that photosystem I and photosystem I1 are segregated in different regions of the membrane; this is in spite of the fact that the glycerolipid phase of the membrane has an unusually low viscosity (2). The majority of PS I1 reaction centers are presently thought to be located in regions of membrane appression, their lateral mobility apparently arrested by membrane surface-to-surface interactions (3). PS I reaction centers are located predominantly in stroma-exposed portions of the membrane (3) excluded from membrane-appressed regions by as yet undefined factors. The cytochrome b6/f complex appears to be fairly uniformly distributed (3) and appears to retain a translational mobility which allows it to diffuse to and from appressed membrane regions. The different placements of PS I and PS 11 reaction centers in the lamellar membrane require that efficient intersystem electron transfer occur over relatively large average distances. Plastoquinol is known to mediate electron transfer between the PS I1 reaction center (4) and the cytochrome b6/f complex and plastocyanin between the cytochrome b6/f complex and the PS I reaction center (5,6). However, there is little hope of understanding the reIative contributions of the mobilities of plastoquinol, plastocyanin, and the cytochrome b6/f complex in catalyzing efficient electron transfer over the necessarily large distances without accurate information about the stoichiometry of the involved components. In addition, this same quantitative information is essential to an understanding of rate limitation in photosynthetic electron transfer. This paper presents the results of experiments designed to measure the stoichiometries of the known carriers relative to the stoichiometries of the PS I and PS I1 reaction centers.
The method employed involved the use of a modified indophenol dye, methyl purple, as a uniquely sensitive and reliable measure of electron transfer through PS I. Since this dye, The abbreviations used are: PS I and PS 11, photosystem I and photosystem II; DBMIB, 2,5-dibromo-3-methyl-6-isopropyl-p-benzoquinone; DCMU, 3-(3',4'-dichlorophenyl)-l,l-dimethylurea; P7m, primary electron donor of the photosystem I reaction center; Q , , primary quinone electron acceptor of the photosystem 11 reaction center. unlike the more commonly used indophenols, accepts electrons exclusively through photosystem I (18), its reduction can be used to measure quite precisely the number of electrons delivered to P700 from the intermediate carriers. Furthermore, the redox state of t h e different carriers can be manipulated by the experimental protocol, and the accessibility of the different carriers to PS I can be varied by inhibitors. Thus the electrons can be drained from various carriers at will and their electron storage capacity determined by measuring the amount of methyl purple reduced. Our findings corroborate those of Whitmarsh and Ort (7) arrived at by an entirely different approach, namely that there is a 1:l:l relationship among the three electron transport complexes in spinach chloroplasts: photosystem I1 reaction centers, cytochrome be/ fcomplexes, and photosystem I reaction centers. Furthermore, we find that the photochemically active plastoquinone pool amounts to nearly 6 molecules/PS I, that the number of reducing equivalents stored in the cytochrome b6/f complex is slightly more than 2 electrons/P700, and that slightly more than 2 molecules of plastocyanin are present for each PS I reaction center.
Chlorophyll Determination-Chlorophyll was measured spectrophotometrically in 80% acetone using the specific absorption coefficients for chlorophyll a at 664 nm and chlorophyll b at 647 nm determined by Ziegler and Egle (9). These two wavelengths corresponded well with the absorption peaks that we measured (0.3-nm half-band width) for chlorophyll a and chlorophyll b purified from spinach (10) and finally solubilized in 80% acetone. The following equations were derived to give the micromolar concentration of chlorophyll a (CJ, chlorophyll b (Cd, and total chlorophyll. to minimize actinic effects of the measuring beam a shutter was placed between the sample and the measuring beam lamp. The sample was illuminated by the measuring beam for 1 s during each measurement. It was verified that neither lower measuring beam intensity nor shorter exposure of the sample to the measuring beam resulted in any change in the state of reduction of the electron transfer chain. The photomultiplier was protected from actinic light by two 590-nm interference filters (Melles Griat, Baird-Atomic, 10-nm half-band width). The amplified signal was digitized using a Biomation 805 transient recorder and the digitized signals summed (Nicolett 1174, Nicolett Instrument C;,rp., Madison, WI) for signal averaging. Measurement of the Photooxidation of Cytochrome f-The number of cytochrome f molecules oxidized by a 200-ms illumination period was measured as described elsewhere (7). The cytochrome f concentration was determined using a reduced minus oxidized molar absorptivity of 20,000 "l.cm" for the wavelength pair 554 minus 540 nm  Actinic Light Sources-Actinic light for long flashes was provided by a 24-V/250-watt tungsten-halogen lamp. The light was passed through a red sharp cutoff filter (Corning 2-58, Corning Glass Works, Corning, NY), and the duration of the light pulse was controlled by an electronic shutter (Uniblitz, Rochester, NY). The light intensity at the sample was 1000 microeinsteins. m-'.s".
Single turnover flashes were provided either by a xenon lamp or by a dye laser. The xenon lamp (model FX-200, Flashtube, E G G , Salem, M A ) emitted a flash (9.7 J total discharge energy) of 6 ps at half-peak height which passed through a sharp red cutoff filter (Corning 2-59). The dye laser (model DL-1000, Phase-R, New Durham, NH) flash was 0.5 ws at half-peak height (43 J total discharge energy). Cresyl violet (50 p~ in methanol) which lases at about 670 nm was circulated through the dye chamber.
Assuming the predicted exponential saturation of the reaction centers as a function of increasing flash intensity, it was calculated from light attenuation experiments with calibrated neutral density filters that both the laser and xenon flashes were more than 98% saturating, i.e. that more than 98% of reaction centers were oxidized by the unattenuated flashes.

RESULTS
An understanding of the rationale of the following experiments is aided by reference to Fig. 1. This figure together with its legend depicts t h e effect of the various experimental protocols described below on the redox state of the individual electron carriers which in turn dictates whether or not a particular carrier is able to donate electrons to methyl purple.

PS Z Reaction Center Content
The PS I reaction center content of the lamellar membranes was determined by measuring the amount of methyl purple reduced in response to a saturating laser flash ( Fig. 2A and Table I). In this experiment poly-L-lysine (290,000 average Mr) was present to inactivate plastocyanin (16) and thereby remove altogether any possibility of rapid reduction of oxi-  shows rapid reduction of methyl purple associated with the depletion of electrons stored in the high-potential carriers between the photosystems followed by a slower steady-state rate associated with the transfer of electrons originating from water. Trace C shows the same rapid reduction associated with the transfer of electrons from the intersystem carriers, but the further reduction is blocked by DCMU.
Note that the DCMU-insensitive electron transfer in trace C comes entirely from the cytochrome be/f complex, plastocyanin, and P7w, since in these dark-adapted membranes the plastoquinone pool is fully oxidized. For Traces B and C each of two samples was flashed four times (150-ms flash) at 30-s intervals, and the results were summed. dized PS I reaction centers and, therefore, any possibility of double turnovers of the reaction centers. Inactivation of plastocyanin by KCN (17) rather than by polylysine gave identical results (data not shown). In the absence of polylysine, a comparison of methyl purple reduction in response to a 6-ps xenon flash and the 500-11s laser flash indicated between 5 and 10% double turnovers of PS I reaction centers were permitted by the longer flash (data not shown). This incidence of double turnovers is consistent with a reduction of PTo0 by    Two equivalents of electrons per cytochrome be/f complex contribute 2.4 mol of electron:mol of P7w to the total rapid donor pool.
6The photochemically active plastoquinone pool is about 10% larger than this value due to the fast and, therefore, undetected loss of about 1 electron/PTW from the pool to the rapid donor chain. The rereduction of the partially oxidized rapid donor chain by the plastoquinol occurs within a few milliseconds following the end of the preillumination (see text). e For the purposes of this table the term Photosystem I1 is defined as PS I1 reaction centers that are able to oxidize water and reduce plastoquinone.
its primary donor with a half-time of about 100 p s . By such experiments it was shown that the PS I reaction center content of lamellar membranes from spinach was 1.57 +-0.10 meq.moI chlorophyll" (Table 11).

Interphotosystem Electron Carriers
Carriers Which Donate Electrons Rapidly to PS I Reaction Centers-Electrons in the higher potential electron carriers on the PS I side of the rate-limiting step of electron transfer can be measured in chloroplast lamellae in which plastoquinone is fully oxidized and PS I1 turnover is prevented by the presence of DCMU or by heat inactivation of water oxidation (see Fig. 1).
We have reported previously ((18) see also Ref. 19) that the light-induced reduction of methyl purple indicates an initial period of rapid electron transfer. This initial period persists for about 20 ms (Fig. 2, truce B ) and is insensitive to DCMU (Fig. 2, trace C). Subsequent to the period of rapid electron transfer, inhibitors of PS I1 turnover abolish any further absorption changes at 591 nm (Fig. 2, truce C ) which indicates that methyl purple is neither further reduced nor reoxidized.
The DCMU-insensitive electron transfer amounts to 8.66 +-0.33 meq.mol chlorophyll" (Table 11) and represents the flow of electrons to methyl purple from the already reduced elec-

4.4).
Quantitation of the rapid donors to P 7 W presupposes that the plastoquinone pool is fully oxidized and that the pool of rapid donors (carriers of the cytochrome ba/f complex and plastocyanin) is fully reduced. In order to establish the fact that the higher potential rapid donors are indeed all reduced the sample can either be preilluminated in the absence of both DCMU and any exogenous acceptor, or ascorbate can be added until there is no further increase in the DCMU-sensitive methyl purple reduction (Fig. 3). The effects of preillumination on the reduction state of the electron transfer chain warrant further comment. Preillumination in the absence of an electron acceptor promotes not only the reduction of the rapid donor pool of P 7 0 0 but also of the plastoquinone pool that is normally fully oxidized at ambient redox potential in dark-adapted lamellar membranes. Indeed, the light-induced reduction of plastoquinone by PS I1 turnover is exploited in experiments described in Fig. 7 to quantitate the size of the active quinone pool. For the data reported in Fig. 3, preillumination was followed by the addition of DCMU and methyl purple and then by a second illumination period sufficient in duration (500 ms) to remove all electrons from the chain subsequent to the site of DCMU inhibition. The rapid donors to P700 are rereduced (t% s 3.5 s) by the photosynthetically reduced methyl purple, but the low potential of plastoquinone prevents its rereduction, creating the situation of a fully oxidized plastoquinone pool and fully reduced rapid donor pool. Ascorbate predictably had no effect on the number of electrons stored by the rapid donor pool in the membranes which had been preilluminated (Fig. 3, o"--o). But in membranes dark adapted under aerobic conditions (A-A) and in membranes heat treated under aerobic conditions (a"0) ascorbate was necessary to restore complete reduction of the rapid donors. However, regardless of the pretreatment, ascorbate (or glutathione, data not shown) restored the rapid donor pool to the same level, just over 8 meq. mol chlorophyll" or about 4.5 electrons in the pool of rapid donors per P~w .
The cytochrome b6/f complex content of the lamellar membranes was determined by measuring the decrease in the size of rapid donor pool which occurred upon addition of reduced DBMIB (which is known to react at the Rieske iron-sulfur component of this complex (20,21)). Fig. 4 shows that the size of the donor pool was decreased by an amount nearly stoichiometric with the amount of P 7 W present (i.e. 1.91 k 0.12 meq.mol chlorophyll", Table 11). Since DBMIB has no effect on the amount of methyl purple reduced by a laser flash (Table I), this effect of DBMIB on the size of the rapid donor pool cannot reflect any action of DBMIB on P 7 W but must rather reflect the abundance of the iron-sulfur-containing cytochrome b6/f complex, which, therefore, must itself have about the same abundance as P700.
It now remains to inquire into the abundance of the other known rapid electron donor, plastocyanin. Since there are about 4.5 electrons in the pool which are rapidly accessible to P7,,,, and about 2.4 (Table 11)  was generally partially oxidized, and in these cases exogenous ascorbate restored the fully reduced state. The P7w content of this chloroplast lamellar membrane preparation was 1.51 meq. mol chlorophyll". shows the effect of the removal of the Reiske FeS center from the rapid donor pool for membranes partially depleted of their plastocyanin (see text). The P7w content of this membrane preparation was 1.57 meq.mol chlorophyll". electrons from the cytochrome b6/f complex to P 7 0 0 . Mild physical disruption of the heat-treated membranes caused by rapid passage through a narrow bore (150-pm) needle resulted in the loss of a component from the rapid donor pool (see Fig.  4). The amount of this component that was lost was often almost equivalent to the P700 content of these membranes. Yet, based on the amount of methyl purple reduced in the response to a laser flash, it was clear that neither the heat treatment nor the subsequent physical disruption procedure caused any decrease in the content of photochemically active PS I reaction centers (Table I). Furthermore, the data in Fig.  4 (lower truce) indicate that heat-treated disrupted membranes had lost none of the component whose oxidation is inhibited by DBMIB, that is no loss of the Rieske FeS center. Thus the component removed from the pool of fast donors by the disruptive treatment of heat-treated membranes can be assigned to an electron transfer component located in the sequence between the cytochrome b6/f complex and PIW. In view of this assignment plastocyanin is the likely candidate, and indeed, we have confirmed that the disruptive treatment does result in a release of plastocyanin to the external medium. Fig. 5 shows the oxidized minus reduced difference spectrum from a sample of heat-treated disrupted chloroplasts after removal of the lamellar membranes. The spectrum shows a major peak near 600 nm and a minor peak around 770 nm which is in excellent agreement with the oxidized minus reduced difference spectrum reported for purified plastocyanin (23). The amount of plastocyanin detected in the supernatant accounted nearly quantitatively (>80%) for the loss to the rapid donor pool observed after disruption of heat-treated membranes. Plastocyanin is found in the external medium only under conditions in which the pool of rapid donors to P700 has been decreased by this procedure of heat treatment followed by disruption. It is important to recognize that the loss of as much as one plastocyanin molecule per P700 does not prevent the rapid (within 30 ms) and complete oxidation of the cytochrome b6/f complex by P7M) and that this reaction remains fully sensitive to poly-L-lysine. FIG. 5. Oxidized minus reduced difference spectrum of plastocyanin released from heat-treated chloroplast lamellar membranes by mild physical disruption. A 2-ml suspension of lamellar membranes (2.3 mM chlorophyll) was heat treated and then physically disrupted by rapid passage through a narrow bore (150pm) syringe needle. The sample was centrifuged at 135,000 X g for 90 min in order to remove the membranes. Further purification was accomplished by passing the sample through a 0.2-pm Millipore filter. Absorption spectra were recorded with a Cary 219 spectrophotometer before and after the addition of 0.5 mM potassium ferricyanide.
Carriers Which Donate Electrons Slowly to PIN (Plastoquinol)-Electrons move more slowly from the lower potential plastoquinol pool to the higher potential rapid donors in a reaction which is both rate determining and energy conserving. As pointed out earlier, the reduction of plastoquinone can be achieved by preillumination of the membranes in the absence of an exogenous electron acceptor and the reduction state of the pool monitored subsequently by measurement of light-induced methyl purple reduction in the presence of DCMU that is added to prevent any further flow of electrons from PS I1 to plastoquinone. Fig. 6 shows the amount of methyl purple reduced by electrons from the intersystem carriers under a variety of conditions. In trace a, all of the carriers had been reduced by preillumination as described above, whereas during a second illumination given to the same sample (trace d) only the higher potential rapid donors remained reduced (by a slow back reaction with the photosynthetically reduced methyl purple). Thus the number of electrons stored in the plastoquinol pool is given by the difference in methyl purple reduction shown in traces a and d of Fig. 6.
Following the initial rapid phase of electron transfer to methyl purple, assigned to oxidation of the cytochrome bs/f complex, plastocyanin, and PIOO itself, the rate of further methyl purple reduction is sensitive to the inhibitors of plastoquinol oxidation 2,4-dinitrophenylether of iodonitrothymol ( Fig. 6, truces b and c) and DBMIB (data not shown).  period (truce a). The DCMU-insensitive methyl purple reduction in truce a is the result of the electrons stored in the rapid donor pool as well as electrons stored as plastoquinol. Truce d represents the DCMU-insensitive methyl purple reduction observed on the third and all subsequent flashes spaced at 60-s intervals. This methyl purple reduction results from electrons stored in the rapid donor pool, electrons which are replenished in the dark intervals by the photosynthetically reduced methyl purple. Thus the difference between truces a and d is a quantitative measure of the number of electrons stored as plastoquinol. The effect of 2,4-dinitrophenylether of iodonitrothymol on light-induced plastoquinol oxidation is shown in truces b (10 pM) and c (20 p~) .
The truces are the averages of one flash to each of two samples.
Plastoquinol undergoes an aerobic oxidation, and this loss of electrons from the pool to oxygen must be accounted for in determining the maximum capacity of the pool to store electrons. Under the conditions of our measurements, the aerobic oxidation of plastoquinol appears first order with a half-time of approximately 60 s (Fig. 7). Extrapolation of the linear semi-log plot back to end of the preillumination period used to reduce plastoquinone shows a pool size of 16.57 f 1.41 meq.mol chlorophyll" (Table 11) or 10 to 11 electrons per P700.
This extrapolation does not take into account any rapid oxidation of plastoquinol which would occur due to the reduction of any of the higher potential PS I components that were in the oxidized state when the preillumination ended. However, the conditions of the preillumination, that is saturating light, the absence of an electron acceptor, and the presence of an uncoupler favor reduction of the entire chain. The lower set of points in Fig. 7 depicts the results obtained when methyl purple was present during the preillumination, a condition which would favor oxidation of the rapid donors to P700. The fact that the extrapolated values (in the presence of an exogenous electron acceptor and its absence) differ by an amount closely approaching the value obtained for the rapid donor pool (7.2 versus 8.7 meq.mo1 chlorophyll") indicates that the rapid donors to P700 shared no more than about one oxidizing equivalent per P700 when the preillumination period in the absence of an exogenous electron acceptor ended. This conclusion is corroborated by measurement of the oxidation state of cytochrome f a few milliseconds after the end of the preillumination. These measurements (data not shown) indicated that with no exogenous acceptor present from 1 to 1.5 oxidizing equivalents were distributed among the rapid donors, predominantly between cytochrome f and the Rieske FeS center, at the end of the preillumination period. The levels of oxidation of the Rieske center and of plastocyanin were calculated (24) based on their equilibrium midpoint potentials, the equilibrium midpoint potential of cytochrome f, and the oxidation state of cytochrome f measured at the end of the preillumination period. Thus a maximum correction of about 10% should be applied to the value for size of the active plastoquinone pool obtained by the extrapolation procedure depicted in Fig. 7 giving a corrected value very close to 6 molecules (11 to 12 electrons) per P700.

The Relative Abundances of PS I and PS II Reaction Centers
The ability to monitor changes in the reduction level of the plastoquinone pool gave us an opportunity to measure the balance between PS I and PS I1 in spinach chloroplast lamellae. Any excess of water-oxidizing PS I1 centers over plastoquinol-oxidizing PS I centers will result in the accumulation of electrons as plastoquinol when all reaction centers are turned over in unison by saturating single turnover flashes. Any accumulation of plastoquinol will be exactly in proportion to the imbalance of PS I1 over PS I. When the chloroplast lamellar membranes were illuminated by single turnover flashes delivered at 5 Hz in the presence of an electron acceptor and an uncoupler, only a very small net reduction of the plastoquinone pool was observed (Fig. 8).
The data in Fig. 8 and Table I1 show only a 5 to 10% excess of PS I1 centers over PS I centers. A doubling of the flash frequency had no effect on these results. The fact that substitution of the very low redox potential acceptor methyl viologen for methyl purple during the preilluminating flash series resulted in no appreciable difference in the measured reduction state of the plastoquinone pool (0) confirms our belief that methyl purple does not oxidize plastoquinol or otherwise intercept electrons prior to photosystem I. Substantial reduction of the plastoquinol pool can be achieved by single turnover flashes if P700 oxidation is impaired by the absence of an electron acceptor (Fig. 8, upper curve). The Reducing equivalents in the plastoquinone pool were assayed by the procedures described in Fig. 6.
initially gentler slope of the no-acceptor condition (A-A) probably results from the presence of the endogenous PS I acceptors. Thereafter (after about 5 turnovers) approximately 0.75 electron/flash accumulates as plastoquinol in the absence of an electron acceptor. An increase in plastoquinol formation from saturating single turnover flashes may also occur when P7,,,, rereduction is impaired by the incomplete reduction of the rapid donor pool as when the absence of an uncoupler limits the oxidation of plastoquinol by the cytochrome b~/f complex (data not shown).

DISCUSSION
The data presented in this paper demonstrate that the pool of high-potential electron carriers which serve as the rapid donors to P7,,,, in spinach lamellar membranes is about 4.5 times larger than the content of P7,,,, itself. This value is substantially larger than earlier reports have suggested (e.g. . In view of the narrow range of values that we observed over the course of this study for commercial spinach obtained in different seasons and from different geographical locations, we do not believe that the origin of the difference lies with inherent differences in plant material. We observed in our own experiments several ways in which misleadingly low estimates of the rapid donor pool size may occur. 1) In chloroplasts incapable of water oxidation (due, for instance, to heat treatment or the addition of DCMU) unanticipated oxidation of the electron carriers can result from actinic effects of the measuring beam. In the experiments reported here the sample was exposed to a weak measuring beam (6 x lo-' J. m-'.s-l, most of which was absorbed by the methyl purple present in the sample) for only 600 ms prior to the actinic light flash. 2) In many preparations of dark-adapted chloroplast lamellar membranes these high-potential electron carriers were, at ambient redox potential, in a partially oxidized state. The fully reduced state can be ensured by preillumination or by the addition of low concentrations of ascorbate as shown by the data of Fig. 3. 3) A decrease in the number of stored reducing equivalents might also result from isolation or from measurement procedures in which a portion of the plastocyanin is lost. These and perhaps other factors could contribute to decrease the apparent size of the rapid donor pool of P7,,,,.
We measured the cytochrome b6/f complex content of the lamellar membrane based on the effect of DBMIB on the DCMU-insensitive methyl purple reduction. The effect was to decrease the amount of methyl purple by an amount just slightly greater than the P7m content of the membrane ( Fig.  4 and  (Fig.  3) and thus argue strongly against any contribution by a bound quinol in the measurements reported here.
We have assigned the remaining 2.1 (Table 11) electrons in the rapid donor pool to plastocyanin. As much as one-half this amount can be released from the membrane vesicles by heat treatment followed by mild physical disruption. The reducing. equivalents lost from the rapid donor pool by this treatment can be nearly quantitatively accounted for by the plastocyanin detected in the supernatant (Fig. 5). The plastocyanin which is resistant to release from within the vesicles is approximately equal to the P7m content and must serve as a mobile electron carrier in subsequent illuminations, since the membranes' entire complement of cytochrome bs/f complex can still be rapidly photooxidized. Inasmuch as plastocyanin is fully reduced under the conditions in which a maximum of one-half of the total can be released from the membranes, it is likely that much of the part retained is bound to the PS I reaction center. Subsequent photooxidation would result in release of the plastocyanin from the reaction center binding site and allow it to serve as a mobile electron shuttle. The method described in Figs. 6 and 7 of this paper introduces a reliable and accurate procedure for the quantitation of the photochemically active plastoquinone pool. The reliability of the method results from the fact that a relatively large absorbance change is measured upon reduction of a dye present in aqueous solution. The molar absorptivity of this exogenous electron acceptor is known with considerable precision, and interpretation of its light-induced absorbance change is straightforward. Our value of nearly 6 active plastoquinones per PTm ( i e . 11 to 12 electrons to P700) is in close agreement to the value reported for spectroscopic measurements in the ultraviolet by Stiehl and Witt (25). These determinations compared the absorbance change at 265 nm induced by a single turnover of PS I1 with the maximal change at 265 nm associated with complete conversion of plastoquinone to plastoquinol. Direct calculation of the plastoquinone pool size based on a differential extinction coefficient for plastoquinone/plastoquinol is complicated by the uncertainties of absorption flattening. When the seemingly appropriate correction is applied (a multiplicative factor of about 1.5 at 265 nm) much larger plastoquinone pool sizes are suggested There has been a controversy regarding the relative abundance of PS I T and PS I in spinach membranes. Recently, Whitmarsh and 01% (7) have challenged the reports of a 2-fold excess of PS I1 centers over PS I centers (33), reports that were based on UV absorption measurements of QA reduction. Whitmarsh and Ort found that the PS 11 content of spinach lamellar membranes estimated by the UV absorption measurement was twice the value given by direct measurement of water oxidation. In this paper we have taken an entirely new approach which confirms the conclusion by Whitmarsh and Ort (7) of a one-to-one relationship between PS I and PS I1 in spinach chloroplast membranes isolated from normally developed plants. We have measured the change in reduction state of the plastoquinone as a function of the number of synchronous turnovers of all reaction centers. Any excess of PS I1 centers over PS I centers would lead to net reduction of the plastoquinone pool whereas the pool would remain fully oxidized if the two photosystems were present in the membranes in equivalent amounts or if PS I was present in excess. Our results indicate only a very slight excess of PS I1 centers (5 to lo%, Table 11) since only 0.05 to 0.1 meq. mol chlorophyll". flash-' was observed to accumulate in the plastoquinone pool when all PS I and PS I1 reaction centers operated in unison. Melis and Anderson (33) cite the net reduction of the plastoquinone pool that they observe under weak green light illumination as qualitative support for a substantial excess of PS I1 reaction centers. Their interpretation of these measurements depends upon optimistic assumptions about the absorption characteristics of each photosystem for the incident green photons, and the uncertainty introduced by these assumptions is difficult to evaluate. In addition the measurements were conducted in the absence of any uncoupler. Our observation that even with low frequency flashes (5 Hz) PS I turnover can be somewhat impaired by incomplete P700 rereduction when an uncoupler was absent leaves little doubt that an uncoupler is necessary to ensure rereduction of P700 in low light experiments as well.
The data presented in this paper demonstrate a remarkable constancy in the stoichiometric relationships of electron transfer components as well as a constancy in the ratio of each component to the total chlorophyll content of the membrane (Table 11). Extreme values were never encountered even though the spinach was obtained from commercial sources originating from widely different places during all seasons of the year. A survey of the literature of the past decade does not reveal the constancy demonstrated in our studies even for this single species. Indeed, the wide range of literature values has been viewed as support for plasticity in electron transfer component ratios and chlorophyll antennae sizes. There is considerable current interest in the extent to which a species of plant can tailor its photosynthetic apparatus to a particular set of environmental conditions and thereby optimize its photosynthetic performance. There is also much current interest in the different strategies adopted by those plant species evolutionarily adapted for extremes in light intensity and spectral distribution. Our study presents a reliable and highly versatile set of techniques ideally suited for such studies. In addition, our results as well as those of Whitmarsh and 01% (7) show the need to re-evaluate some of the conclusions regarding capacity of plants to manipulate their photosynthetic apparatus, particularly those conclusions which are based on the wide range of numbers and component ratios in the literature.