Plant Cell Physiol. 42(8): 842–848 (2001)JSPP © 2001

Inhibition of Photosystems I and II and Enhanced Back Flow of PhotosystemI Electrons in Cucumber Leaf Discs Chilled in the Light

Sun-Ju Kim 1, 3, Choon-Hwan Lee 1, 3, Alexander B. Hope 1, 2 and Wah Soon Chow 1, 4

1 Research School of Biological Sciences, Australian National University, GPO Box 475, Canberra, ACT 2601, Australia2 School of Biological Sciences, Flinders University, GPO Box 2100, Adelaide, SA 5001, Australia

;Pre-illumination of cucumber leaf discs at a chilling

temperature in low-irradiance white light resulted in accel-erated re-reduction of P700+ [the special Chl pair in thephotosystem I (PSI) reaction centre] when the far-red mea-suring light was turned off. Measurements (in ���� methylviologen or ���� DCMU conditions) of the re-reduction halftime suggest that accelerated re-reduction of P700+

appeared to be predominantly due to charge recombina-tion and only partly due to reductants sustained by previ-ous cyclic electron flow around PSI. Apparently, chargerecombination in PSI was greatly enhanced by inhibition offorward, linear electron flow. Inhibition of PSII electrontransport was observed to occur to a lesser extent than thatof PSI, but only if the measurement of PSII functionalitywas free from complications due to downstream accumula-tion of electrons in pools. We suggest that promotion of con-trolled charge recombination and cyclic electron flow roundPSI during chilling of leaves in the light may partly pre-vent further damage to both photosystems.

Key words: Chilling — Cucumis sativus L. — Charge recom-bination — Cyclic electron transport — Photosystem I — Pho-tosystem II.

Abbreviations: DBMIB, 2,5-dibromo-3-methyl-6-isopropyl-p-benzoquinone; Fo, Fm, minimum and maximum Chl fluorescence yieldcorresponding to open and closed PSII, respectively, after dark pre-treatment; Fo�, Fm�, minimum and maximum Chl fluorescence yieldcorresponding to open and closed PSII during illumination, respec-tively; Fv = Fm – Fo; NDH, NAD(P)H dehydrogenase; P700, specialChl pair in the PSI reaction centre; PQ, plastoquinone; QA, primaryquinone acceptor in PSII; qp, photochemical quenching coefficient.

Introduction

Cucumber leaves are readily photodamaged in chillingconditions under low light, the main site of damage being pho-tosystem I (PSI) rather than PSII (Terashima et al. 1994,Sonoike 1996a, Sonoike 1999). Photodamage to PSI involvesthe destruction of iron-sulfur (Fe-S) centres Fx, FA and FB

(Sonoike et al. 1995) and degradation of the psaB gene prod-uct, due to the generation of reactive oxygen species (Sonoike

1996b). Even field-grown, chilling-tolerant barley, whenexposed to bright daylight in combination with chilling temper-atures, experiences a certain extent of PSI photodamage asmanifested by damage to the Fe-S centres and to the PSI-A/Bpolypeptides (Teicher et al. 2000). The above studies demon-strate that photoinhibition of PSI is a physiologically relevantphenomenon.

As a consequence of perturbations of the PSI reaction cen-tre on chilling cucumber leaves in the light, normal linear elec-tron transfer is retarded, leading to enhanced charge recombi-nation as indicated by an abundance of triplet states of Chl(Sonoike et al. 1995); this is also proposed for chilling ofpotato leaves (Havaux and Davaud 1994). Another conse-quence has been suggested to be the enhancement of cyclicelectron flow around PSI (Sonoike 1998, Sonoike 1999). Botheffects are expected to contribute to faster re-reduction ofP700+, but their relative contributions are difficult to separate.In this study, we have made an initial attempt to rank the rela-tive contributions of the two paths of electron donation toP700. Our study also monitors the functionality of PSII, inhib-ited to a lesser, but not negligible, extent compared with PSI. Ascrutiny of methods of monitoring PSII functionality is neces-sary because downstream accumulation of electrons in poolsmay complicate the quantification of PSII units.

Results and Discussion

Different modes of monitoring light-induced oxidation of P700P700, the special Chl pair of PSI, undergoes photooxida-

tion in the light, and re-reduction in the dark. Fig. 1a showsthat measurements in single wavelength, dual wavelength,reflectance or transmittance modes gave essentially the samenormalized signals. The best signal-to-noise ratio and signalstability was obtained in the dual wavelength, reflectancemode, as the example in Fig. 1b shows.

Inhibition of photosystem IPre-illumination of cucumber leaf discs at 4�C markedly

suppressed the extent of photooxidation of P700 by far-redlight (170 �mol m–2 s–1, transmitted by a 1 mm thick RG 9filter) given at room temperature, and measured at about 2 minafter pre-illumination at 4�C. Thus, after 3 h pre-illuminationwith white light at 160 �mol m–2 s–1 and 4�C, the extent of pho-

3 Permanent address: Department of Molecular Biology, Pusan National University, Pusan, 609-735 Korea.4 Corresponding author: E-mail, chow@rsbs.anu.edu.au; Fax, +61-2-6125-8056.

842

Electron flow after chilling cucumber in light 843

tooxidation of P700 by far-red light was decreased by an orderof magnitude (solid circles, Fig. 2a). In an attempt to ensuremaximum photooxidation of P700, a strong pulse of white lightwas added to the far-red light being given to leaves (Siebke etal. 1997). The total photooxidation of P700 then possibly rep-resented the full extent of photooxidisable P700. As can be

seen in Fig. 2a (open circles), the putative full extent of pho-tooxidisable P700 also declined, by a factor of 3 in 3 h.

The above measurements of P700 photooxidation were per-formed 2 min after taking the leaf discs out of the chilling treat-ment. When this relaxation time was varied, it was found thatthe effect of chilling in the light persisted in the dark at room

Fig. 1 (a) The dependence of the relative signal associated with P700 photooxidation (in a control cucumber leaf attached to the plant) on thefar-red irradiance. Far-red irradiance (transmitted by an RG715 long-pass filter) was varied by adjusting the aperture of the shutter, and given tothe leaf until a steady signal was reached. Signals obtained in the different modes were normalized to the respective maxima at 10.8 �mol m–2 s–1.(b) Typical signals of P700 photooxidation induced by far-red light (10 s, 40 �mol m–2 s–1, transmitted by a 1 mm-thick RG9 long-pass filter)given to leaf discs. The signals were obtained in the reflectance dual-wavelength mode in the absence or presence of methyl viologen, followingpre-illumination (160 �mol m–2 s–1) of leaf discs at 6�C for 1 h.

Fig. 2 (a) Decline in the P700 photooxidation signal (reflectance mode, single wavelength), measured at room temperature, as a function of timeof pre-illumination (160 �mol m–2 s–1) of leaf discs at 4�C. Photooxidation of P700 was induced by either far-red light alone (closed circles, dura-tion 20 s, 170 �mol m–2 s–1) or a combination of 20 s far-red light to which was added a pulse of intense white light (open circles, duration 1 s,6,500 �mol m–2 s–1) at about 19.5 s from the beginning of far-red illumination. Both types of signals were recorded at about 2 min after the cessa-tion of chilling treatment in the light. (b) The extent of P700 photooxidation (reflectance mode, single wavelength) as a function of time of darkrelaxation at room temperature following chilling in the light (4 h at 4�C, 160 �mol m–2 s–1). Symbols as in (a).

Electron flow after chilling cucumber in light844

temperature for min, after which the P700 photooxidation signal(induced either by far-red light alone or far-red plus a strongpulse of white light) recovered somewhat in an h (Fig. 2b).

Inhibition of photosystem IIIllumination of cucumber leaf discs at 4�C led to a slight

decrease in Fv/Fm, the ratio of variable to maximum Chl fluo-rescence yields representing the optimum quantum efficiencyof photochemistry in PSII after a dark treatment for 30 min.The slight decline in this ratio implies an apparently small lossof optimum quantum efficiency as a result of chilling in thelight (Fig. 3, open circles). However, Fv/Fm may sometimesgive a false impression of a limited effect. For example, in pealeaf discs, photoinactivation of the first 30% of PSII units hadlittle effect on Fv/Fm (Park et al. 1996).

Generally, the oxygen yield per single-turnover flash,obtained in repetitive flash illumination, gives a direct measureof functional PSII units in leaf discs (Chow et al. 1989, Chowet al. 1991). In the case of severe, concurrent inhibition of PSI,however, the oxygen yield could underestimate the number offunctional PSII units. This is because when functional PSIIunits are much more abundant than functional PSI units, moreelectrons are delivered by each single-turnover flash to theplastoquinone pool than can be removed via PSI (Chow et al.1989); then, in repetitive flashes, increased reduction of the PQpool could keep QA in some PSII units reduced via chemicalequilibrium. PSII units with reduced QA, otherwise functional,are consequently unable to evolve oxygen on the next flash.This is the most likely explanation for the apparent number of

functional PSII units declining markedly, by about 80% in 4 h,as a result of chilling in the light (Fig 3, solid circles).

To substantiate this explanation, we examined the reduc-

Fig. 4 The Chl fluorescence yield F during repetitive (10 Hz) flash illumination of leaf discs in the presence of far-red (FR) light (~1 �mol m–2

s–1). Flash illumination (FL) started and ended as indicated by the upward and downward arrows, respectively. The photochemical quenchingcoefficient qp during flash illumination was calculated as (Fm� – F)/ (Fm� – Fo�) (Schreiber et al. 1994). The value of qp was 0.99 for a control leafdisc (left) and 0.53 for a leaf that had been illuminated at 180 �mol m–2 s–1 for 5.5 h at 5�C. The plant materials used in this experiment and that inFig. 3 were grown about 2 years apart, and hence may be only qualitatively comparable.

Fig. 3 Decline in the functionality of PSII, as indicated by the Chlfluorescence parameters Fv/Fm and 1/Fo – 1/Fm, and the apparent oxy-gen yield per single-turnover flash, as a function of time of pre-illumination (160 �mol m–2 s–1) at 4�C. At zero time of pre-illumination,the mean control values were 0.829 for both fluorescence parameters,and 3.52 mmol PSII (mol Chl) –1 for functional PSII units (assumingthat each functional PSII evolves one O2 molecule every four flashes).The dotted line is an estimation of the true time course of decline offunctional PSII complexes (see text).

Electron flow after chilling cucumber in light 845

tion state of QA during repetitive flash illumination. The reduc-tion state is measured as the photochemical quenching coeffi-cient qp (Schreiber et al. 1994). As shown in Fig. 4, a controlleaf disc, repetitively illuminated by flashes at 10 Hz as inmeasurements of O2 evolution, had a qp value of 0.99. Thus,almost all the traps were open, and the oxygen yield per singleturnover flash should have given a good estimate of functionalPSII content. In contrast, qp was 0.53 in a leaf disc that hadbeen illuminated at 5�C for 5.5 h, implying that many PSIItraps were closed and, therefore, would not have yielded oxy-gen on repetitive flash illumination. Thus, the oxygen yield persingle turnover flash underestimated the functional PSII con-tent in leaf discs in which PSI was severely photoinactivated.

Therefore, we examined an alternative measure of func-tional PSII units. A more reliable measure of functional PSIIunits under the circumstances may be the Chl fluorescenceparameter, 1/Fo – 1/Fm. This parameter has been suggested toreflect changes in reaction centre function (Havaux et al. 1991,Walters and Horton 1993), and is linearly correlated with theoxygen yield per flash under conditions where no inhibition ofPSI occurs (Park et al. 1995, Lee et al. 1999). In Fig. 3 (opendiamonds), 1/Fo – 1/Fm declined more slowly than the oxygenyield per flash but faster than Fv/Fm. The parameter 1/Fo – 1/Fm

possibly best reflects the content of functional PSII units.To further test 1/Fo – 1/Fm as a valid measure of func-

tional PSII content, we illuminated leaf discs at 25�C at whichno photoinactivation of PSI takes place (Terashima et al. 1994).Under such circumstances, the oxygen yield per single turno-ver flash gives a valid measure of functional PSII content. Fig.5 shows a linear correlation between 1/Fo – 1/Fm and func-

tional PSII content. The straight lines did not pass through theorigin, however. In previous reports, the straight line relation-ship did not pass through the origin in pea leaves (Park et al.1995), but passed nearly through the origin in Capsicum leaves(Lee et al. 1999). Assuming that the same straight line relation-ship still holds after chilling in the light, then one can estimatethe functional PSII content from 1/Fo – 1/Fm, as given by thedotted line in Fig. 3. Notably, the decline in functional PSII, asindicated by the dotted line in Fig. 3, was slower than that ofphotooxidisable P700 (compare Fig. 2a and Fig. 3). This obser-vation is consistent with the reported preferential inhibition ofPSI during chilling in low light (Terashima et al. 1994, Sonoike1996a, Sonoike 1999).

Charge recombination in PSI vs cyclic electron flowIt has been suggested that the capacity for cyclic electron

flow may be enhanced after chilling cucumber leaves in thelight (Sonoike 1998, Sonoike 1999). This suggestion is basedon a slower photooxidation of P700 by far-red light, and afaster re-reduction after far-red light is turned off, assumingthat electrons available for P700 re-reduction come solely fromprevious cyclic electron flow. Accelerated cyclic electron trans-port has also been suggested in a short report to be an earlyeffect of chilling in chilling-sensitive plants (Herbert et al.1997).

On the other hand, when redox components are damagedin PSI after chilling in the light and forward electron flow ishindered, charge recombination is expected to be enhanced,just as the removal or pre-reduction of downstream redox com-ponents accelerates charge recombination (Brettel 1997). Byinfiltration of leaf discs with methyl viologen or with DCMUafter chilling in the light, we hoped to shed some light on therelative importance of cyclic electron flow and charge recombi-nation in the kinetics of P700 re-reduction. Separate measure-ments indicated that infiltration of cucumber leaf discs with30 �M DCMU inhibited the light- and CO2-saturated rate ofoxygen evolution by 98% (data not shown).

In Fig. 6a, cucumber leaf discs were illuminated for vari-ous durations at 6�C, and then infiltrated with either water or asolution of methyl viologen (to diminish cyclic electron flowand charge recombination). The excess water in the leaf tissuewas allowed to evaporate off in the dark for up to 20 min atroom temperature before measurement of P700 photooxidationwith far-red light. On turning off far-red light with an electronicshutter, the half-time of the decay (t1/2) of the absorbancechange was taken to reflect the rate of re-reduction of P700+.Any reductants that have accumulated from previous cyclicelectron flow that persist for some time after cessation of far-red illumination, as well as any charge recombination in PSI,will contribute to faster re-reduction of P700. In the absence ofany chilling treatment, t1/2 was similar for infiltration withwater and methyl viologen, indicating that both processes areprobably negligible in control leaf discs. However, a differencein t1/2 between infiltration with water and methyl viologen soon

Fig. 5 Correlation of 1/Fo – 1/Fm with functional PSII content (deter-mined from the oxygen yield per flash) in leaf discs pre-illuminated at1,500 �mol m–2 s–1 and 25�C for various durations, in the absence orpresence of lincomycin, an inhibitor of chloroplast-encoded proteinsynthesis. At this temperature, there was no appreciable photoinactiva-tion of PSI, so a valid measure of functional PSII content was obtainedfrom the oxygen yield per flash.

Electron flow after chilling cucumber in light846

developed with the pre-treatment of chilling in the light; thedifference persisted with increase in time of pre-illumination at6�C (Fig. 6a). This difference in t1/2 could be attributable to theready availability of reductants sustained in previous cyclicelectron flow in the absence, but not the presence, of methylviologen. Other studies in vitro have also shown that inhibitionof cyclic electron flow by antimycin increased the t1/2 of post-illumination re-reduction of P700+ (Scheller 1996) and, equiva-lently, oxidation of reduced ferredoxin (Cleland and Bendall1992). However, charge recombination, which is diminished byinfiltration with methyl viologen but not water, cannot be ruledout. Indeed, charge recombination in PSI is considered to bepartly responsible for the DBMIB-insensitive reduction ofP700+ (Scheller 1996).

When control leaf discs were vacuum infiltrated withDCMU to block any electron flow from PSII, the t1/2 of post-illumination re-reduction of P700+ was greatly increased: ca.3 s in the presence of DCMU (Fig. 6b) compared with ca. 1 s inthe absence of DCMU (Fig. 6a). The faster re-reduction ofP700+ in the absence of DCMU is attributable to electrons(released during water oxidation by PSII) becoming availablefor rapid re-reduction of P700+. Apparently, even in far-redlight, PSII was able to transfer some electrons to PSI. Conceiv-ably, such PSII electrons could contribute to poising of theinter-photosystem electron carriers in favour of cyclic electrontransport (Arnon and Chain 1979).

However, with progressive photoinhibition of PSI inchilling conditions, the difference between methyl viologenand water persisted even in the presence of DCMU, whichshould have inhibited any poising of cyclic electron transport.

Therefore, we conclude that there was probably little poising ofcyclic electron transport in the absence of DCMU in the firstplace. Rather, the observation that the difference in t1/2 (�methyl viologen) was only slightly diminished by DCMU (Fig.6b) compared with the absence of DCMU (Fig. 6a) indicatesthat charge recombination was the predominant factor respon-sible for the accelerated re-reduction of P700+ after pre-illumination in chilling conditions, at least in short-term chill-ing treatment in the light. In the longer term, acclimation mayoccur to promote cyclic electron flow by increasing compo-nents such as the NDH complex (Teicher et al. 2000).

In the presence of both DCMU and methyl viologen, t1/2

declined steadily with pre-illumination in chilling conditions(Fig. 6b). This can be interpreted as being due to (1) a progres-sive loss of ability of PS I electron acceptors (e.g. the Fe-S cen-tres FA/FB) to donate electrons to methyl viologen, coupledwith (2) a progressive enhancement of charge recombination,for example, in the pairs P700+ FA

– and/or P700+ FB–, leading

to accelerated re-reduction of P700+. While it is known thatcharge recombination between P700+ and FA

–/FB– occurs with a

half-time �50 ms (Brettel 1997), this half-time cannot bedirectly equated with t1/2 in Fig. 6b. This is because there wereprobably two populations of PSI complexes: one which wasable to reduce methyl viologen (but had very limited chargerecombination) and another which was unable to reduce methylviologen (but readily carried out charge recombination). The t1/2

values in Fig. 6b probably represent crude overall half-timesfor the re-reduction of P700+ in the mixed population. Obvi-ously, a detailed analysis of the kinetics of P700+ under a vari-ety of conditions is necessary in order to unravel the various

Fig. 6 (a) The half-time of re-reduction of P700+ in leaf discs on cessation of far-red illumination (40 �mol m–2 s–1, transmitted by an RG 9 fil-ter) as a function of time of pre-illumination (160 �mol m–2 s–1) at 6�C. A 20-min dark period was allowed between the chilling treatment andmeasurement, during which leaf discs were infiltrated with either methyl viologen or water, and excess intercellular water allowed to evaporate.(b) Same protocol as in (a), except that DCMU (30 �M) was also present. Signals in (a) and (b) were obtained in the reflectance, dual-wavelength(810/860 nm) mode.

Electron flow after chilling cucumber in light 847

processes that may contribute to the re-reduction of P700+.The diminished amount of photo-oxidisable P700, due

largely to the accelerated re-reduction of P700+, persisted forsome time even when the leaf discs were returned to room tem-perature (Fig. 2b). Some recovery of the signal, however, wasevident over an h. Perhaps the chilling treatment in combina-tion with light brought about membrane changes (possibly lipidphase changes or lateral migration of membrane protein com-plexes) that took some time to reverse when returned to roomtemperature. Perhaps, too, charge recombination became some-what less likely when the membrane changes were reversed atroom temperature.

Controlled back flow of electrons in/around PSI, whetherby charge recombination or cyclic electron flow under chillingconditions, may help to protect PSI from further damage. In thecase of the former, charge recombination to give the groundstate directly would be an effective way of dissipating excita-tion energy in PSI, as has been suggested by Warren et al.(1993). Even if charge recombination gives rise to the tripletexcited state of P700, no damage would occur if: (1) �-carotenes dissipate triplet P700; (2) triplet P700 is inaccessibleto oxygen (Sétif et al. 1981) so that reactive singlet oxygen isnot formed; or (3) �-carotenes convert any singlet oxygen, ifformed, to ordinary triplet oxygen. In the case of enhancedcyclic electron flow, which may occur in long-term acclima-tion to chilling (Teicher et al. 2000), PSI may be protected inone or two ways. (1) Cyclic electron flow generates a transthyl-akoid pH gradient, which then down-regulates PSII, therebylimiting the flux of electrons along the linear chain (Heber andWalker 1992, Sonoike 1996a). However, this possibility maybe remote, given that uncoupling is likely to occur in chillingconditions (Terashima et al. 1991) and that little zeaxanthin isformed (S.-J. Kim et al., data not shown) presumably due to aninsufficient transthylakoid pH gradient. (2) Cyclic electronflow directs the flow of electrons so that the chance of electrondonation to O2 or, worse still, to H2O2 (to form highly reactivehydroxyl free radicals) is restricted (Chow and Hope 1998).

In summary, based on the results presented, the followingconclusions can be made: (1) chilling of cucumber leaf discs inthe light severely decreased the extent of P700 photooxidationby far-red light, with or without an additional pulse of strongwhite light; (2) the decrease in photooxidisable P700 wasmainly due to charge recombination which accelerated the re-reduction of P700+; (3) electrons accumulated during previouscyclic electron flow probably only facilitated the re-reductionof P700+ to a limited extent; and (4) under conditions wherePSI is inhibited more than PSII, the accumulation of electronsin the PQ pool may give rise to an underestimation of func-tional PSII units assayed by the oxygen yield per repetitiveflash.

Materials and Methods

Plant material, chilling treatment and vacuum infiltrationLeaves were harvested from cucumber (Cucumis sativus L. cv.

Black Magic) grown in a potting mixture. The growth cabinet wasmaintained at a regimen of 24/21�C (day/night) with a 12 h photope-riod. The growth irradiance was 230 �mol m–2 s–1, supplied by a com-bination of metal halide (MBID250/T/H, Kolorac, Hungary) andincandescent lamps. In the chilling treatment, leaf discs were floatedon distilled water maintained at a specified temperature and illumi-nated under fluorescent light (160 �mol m–2 s–1) in a cold room for upto 5.5 h.

When required, leaf discs were vacuum infiltrated with eithermethyl viologen (300 �M), water or DCMU (30 �M) by submergingdiscs in the aqueous phase inside a vacuum desiccator at 0.5 atmos-phere of negative pressure for about 1 min. Excess intercellular waterwas allowed to evaporate from the tissue in the dark for 15–20 minwhich also allowed time for methyl viologen to permeate cell andorganelle membranes.

Measurement of P700P700 photooxidation and re-reduction were measured with a

pulse-amplitude-modulation fluorometer (Walz, Effeltrich, Germany)equipped with either a single wavelength (820 nm) or dual wave-length (810/860 nm) emitter unit, and either in the reflectance mode ortransmission mode, as specified in the figure legends. To test for anypossible inconsistencies between these different modes of measure-ment, a leaf attached to a plant was illuminated with far-red light(transmitted by a 4-mm-thick RG715 long-pass filter) until a steadyP700 oxidation state was reached; the extent of P700 photooxidationwas determined as a function of far-red irradiance, and normalizedaccording to the maximum observed at 10.8 �mol m–2 s–1 (measuredwith a SKP 215 sensor, Skye Instruments Ltd, Llandrindod Wells,Wales). As seen in Fig. 1a, the different modes of measurement gaveessentially the same results. However, the best signal-to-noise ratioand signal stability was obtained with the dual-wavelength-reflectancemode; an example of such signals is given in Fig. 1b.

To record rapid redox kinetics of P700, far-red light was passedthrough an electronic shutter for a set duration (generally 8 s), and thesignal changes were stored in a digital oscilloscope (Gould 1421,Essex, U.K.) for subsequent plotting on a chart recorder. The half-timefor re-reduction of P700 was taken as the time for the P700+ signal todecay by half after cessation of far-red illumination.

Measurement of functional PSII by oxygen yield per flashEstimations of O2 evolution per single-turnover flash, which

were used to monitor the effective concentration of functional PSIIunits, were obtained from measurements with a leaf disc oxygen elec-trode (Hansatech, King’s Lynn, U.K.) and repetitive flash illuminationwith background far-red light (~1 �mol m–2 s–1), as described earlier(Chow et al. 1989, Chow et al. 1991). The estimations are based on theassumption that each functional PSII unit evolved one O2 moleculeafter four flashes.

Measurement of Chl fluorescence yieldsAfter leaf discs were transferred from the chilling treatment in

the light, they were dark-adapted for 30 min at room temperature.Then Fo and Fm (minimum and maximum fluorescence yields corre-sponding to open and closed reaction centres, respectively) were meas-ured with a Plant Efficiency Analyser (Hansatech, King’s Lynn, U.K.)operated at 80% maximum excitation light intensity. To use 1/Fo – 1/Fm as a measure of functional PSII units (see results), all fluorescenceyields were normalized to the mean Fo value of controls (= 1.00).

Electron flow after chilling cucumber in light848

To determine the photochemical quenching coefficient duringrepetitive flash illumination, a pulse amplitude modulation (PAM)fluorometer (Walz, Effeltrich, Germany) was used. A leaf disc wassupported horizontally, adaxial side up, by a UV-absorbing filter inopen air. Fo was measured by supplying weak, modulated red lightdownwards to the adaxial surface of the leaf disc via one branch of amultifurcated light guide. Then continuous far-red light (�1 �mol m–2

s–1) was added via another branch to simulate far-red background illu-mination in the oxygen electrode chamber used to measure functionalPSII by oxygen yield per flash. Fm was measured admitting a 1 s satu-rating pulse of white light (8,000 �mol m–2 s–1) through a third branchwhile the far-red light was being applied. Subsequently, repetitive flashillumination (10 Hz) was applied upwards through a UV-absorbing fil-ter (Schott 115 Tempax plus a perspex sheet) to the abaxial side of theleaf disc. At steady state, the fluorescence yield F was recorded, andFm� was measured using a 1 s saturating pulse of white light. Soonafterwards, flash illumination was turned off to enable Fo� to be meas-ured. The photochemical quenching coefficient qp was calculated as(Fm� – F)/(Fm� – Fo�) (Schreiber et al. 1994).

Acknowledgements

Kim and Lee gratefully acknowledge support from the interdisci-plinary research program of the KOSEF (Grant No. 1999-2-203-001-3) in Korea for this work, and Hope is grateful for a Visiting Fellow-ship from the Research School of Biological Sciences, AustralianNational University.

References

Arnon, D.I. and Chain, R.K. (1979) Regulatory electron transport pathways incyclic photophosphorylation. FEBS Lett. 102: 133–138.

Brettel, K. (1997) Electron transfer and arrangement of the redox cofactors inphotosystem I. Biochim. Biophys. Acta 1318: 322–373.

Chow, W.S. and Hope, A.B. (1998) The electrochromic signal, redox reactionsin the cytochrome bf complex and photosystem functionality in photoinhib-ited tobacco leaf segments. Aust. J. Plant Physiol. 25: 775–784.

Chow, W.S., Hope, A.B. and Anderson, J.M. (1989) Oxygen yield per flashfrom leaf discs quantifies photosystem II. Biochim. Biophys. Acta 973: 105–108.

Chow, W.S., Hope, A.B. and Anderson, J.M. (1991) Further studies on quantify-ing photosystem II in vivo by flash-induced oxygen yield from leaf discs.Aust. J. Plant Physiol. 18: 397–410.

Cleland, R.E. and Bendall, D.S. (1992) Photosystem I cyclic electron transport:Measurement of ferredoxin-plastoquinone reductase activity. Photosynth.Res. 34: 409–418.

Havaux, M. and Davaud, A. (1994) Photoinhibition of photosynthesis in chilledpotato leaves is not correlated with a loss of photosystem II activity. Preferen-tial inactivation of photosystem I. Photosynth. Res. 40: 75–92.

Havaux, M., Strasser, R.J. and Grippin, H. (1991) A theoretical and experimen-tal analysis of the qP and qN coefficients of Chl fluorescence quenching andtheir relation to photochemical and nonphotochemical events. Photosynth.Res. 27: 41–45.

Heber, U. and Walker, D.A. (1992) Concerning a dual function of coupled

cyclic electron transport in leaves. Plant Physiol. 100: 1621–1626.Herbert, S.K., Sacksteder, C. and Kramer, D.M. (1997) Accelerated cyclic elec-

tron transport is an early effect of chilling in chilling-sensitive plants, Ann.Meet. Am. Soc. Plant Physiologists and Can. Soc. Plant Physiologists, Van-couver, Plant Physiol. Suppl. 114: 219.

Lee, H.-Y., Chow, W.S. and Hong, Y.-N. (1999) Photoinactivation of photosys-tem II in leaves of Capsicum annuum. Physiol. Plant. 105: 377–384.

Park, Y.-I., Anderson, J.M. and Chow, W.S. (1996) Photoinactivation of func-tional photosystem II and D1-protein synthesis in vivo are independent of themodulation of the photosynthetic apparatus by growth irradiance. Planta 198:300–309.

Park, Y.-I., Chow, W.S. and Anderson, J.M. (1995) Light inactivation of func-tional photosystem II in leaves of peas grown in moderate light depends onphoton exposure. Planta 196: 401–411.

Scheller, H.V. (1996) In vitro cyclic electron transport in barley thylakoids fol-lows two independent pathways. Plant Physiol. 110: 187–194.

Schreiber, U., Bilger, W. and Neubauer, C. (1994) Chl fluorescence as a nonin-trusive indicator for rapid assessment of in vivo photosynthesis. In Ecophysi-ology of Photosynthesis. Edited by Schulze, E.-D and Caldwell, M.M. pp.49–70. Springer-Verlag, Berlin.

Sétif, P., Hervo, G. and Mathis, P. (1981) Flash-induced absorption changes inphotosystem I radical pair or triplet formation? Biochim. Biophys. Acta 638:257–267.

Siebke, K., von Caemmerer, S., Badger, M. and Furbank, R.T. (1997) Express-ing an RbcS antisense gene in transgenic Flavaria tridentis leads to anincreased quantum requirement for CO2 fixed in photosystems I and II. PlantPhysiol. 115: 1163–1174.

Sonoike, K. (1996a) Photoinhibition of photosystem I: its physiological signifi-cance in the chilling sensitivity of plants. Plant Cell Physiol. 37: 239–247.

Sonoike, K. (1996b) Degradation of a psaB gene product, the reaction centersubunit of photosystem I, is caused during photoinhibition of photosystem I:possible involvement of active oxygen species. Plant Sci. 115: 157–164.

Sonoike, K. (1998) Photoinhibition of photosystem I in chilling sensitive plantsdetermined in vivo and in vitro. In Photosynthesis: Mechanisms and Effects.Edited by Garab, G. Vol. III, pp. 2217–2220. Kluwer Academic Publishers,Dordrecht.

Sonoike, K. (1999) The different roles of chilling temperatures in the photoinhi-bition of photosystem I and photosystem II. J. Photochem. Photobiol. B 48:136–141.

Sonoike, K., Terashima, I., Iwaki, M. and Itoh, S. (1995) Destruction of photo-system I iron-sulfur centers in leaves of Cucumis sativus L. by weak illumina-tion at chilling temperatures. FEBS Lett. 362: 235–238.

Teicher, H.B., Møller, B.L. and Scheller, H.V. (2000) Photoinhibition of photo-system I in field-grown barley (Hordeum vulgare L.): induction, recovery andacclimation. Photosynth. Res. 64: 53–61.

Terashima, I., Funayama, S. and Sonoike, K. (1994) The site of photoinhibitionin leaves of Cucumis sativus L. at low temperature is photosystem I, not pho-tosystem II. Planta 193: 300–306.

Terashima, I., Sonoike, K., Kawazu, T. and Katoh, S. (1991) Exposure of leavesof Cucumis sativus L. to low temperatures in the light causes uncoupling ofthylakoids. II. Non-destructive measurements with intact leaves. Plant CellPhysiol. 32: 1275–1283.

Walters, R.G. and Horton, P. (1993) Theoretical assessment of alternative mech-anisms for non-photochemical quenching of PSII fluorescence in barleyleaves. Photosynth. Res. 36: 119–139.

Warren, P.V., Golbeck, J.H. and Warden, J.T. (1993) Charge recombinationbetween P700+ and A1– occurs directly to the ground state of P700 in a Pho-tosystem I devoid of FX, FB and FA. Biochemistry 32: 4411–4419.

(Received December 14, 2000; Accepted May 27, 2001)