Measurements of variable chlorophyll £uorescence using fast repetitionrate techniques: de¢ning methodology and experimental protocols

Zbigniew S. Kolber, Ondr­ej Pras­il 1, Paul G. Falkowski *Environmental Biophysics and Molecular Biology Program, Rutgers University, 71 Dudley Rd, New Brunswick, NJ 08901-8521, USA

Received 12 December 1997; revised 15 June 1998; accepted 23 June 1998

Abstract

We present a methodology, called fast repetition rate (FRR) fluorescence, that measures the functional absorption cross-section (cPS II) of Photosystem II (PS II), energy transfer between PS II units (p), photochemical and nonphotochemicalquenching of chlorophyll fluorescence, and the kinetics of electron transfer on the acceptor side of PS II. The FRRfluorescence technique applies a sequence of subsaturating excitation pulses (`flashlets') at microsecond intervals to inducefluorescence transients. This approach is extremely flexible and allows the generation of both single-turnover (ST) andmultiple-turnover (MT) flashes. Using a combination of ST and MT flashes, we investigated the effect of excitation protocolson the measured fluorescence parameters. The maximum fluorescence yield induced by an ST flash applied shortly (10 Ws to5 ms) following an MT flash increased to a level comparable to that of an MT flash, while the functional absorption cross-section decreased by about 40%. We interpret this phenomenon as evidence that an MT flash induces an increase in thefluorescence-rate constant, concomitant with a decrease in the photosynthetic-rate constant in PS II reaction centers. Thesimultaneous measurements of cPS II, p, and the kinetics of Q3

A reoxidation, which can be derived only from a combination ofST and MT flash fluorescence transients, permits robust characterization of the processes of photosynthetic energy-conversion. ß 1998 Elsevier Science B.V. All rights reserved.

1. Introduction

Because its measurement is non-destructive, non-invasive, rapid, sensitive, and achieved in real-time

[1], the change in the quantum yield of chlorophyll£uorescence induced by actinic light is used exten-sively to derive photosynthetic parameters relatedto PS II [2]. Theoretical and empirical models have

0005-2728 / 98 / $ ^ see front matter ß 1998 Elsevier Science B.V. All rights reserved.PII: S 0 0 0 5 - 2 7 2 8 ( 9 8 ) 0 0 1 3 5 - 2

Abbreviations: aPS II, optical cross section of PS II; Chl, chlorophyll ; C(t), fraction of closed PS II reaction centers at time t duringFRR excitation protocol; f(t), £uorescence yield at time t during FRR protocol; FRR, fast repetition rate; Fo, minimal £uorescenceyield; Fm, maximal £uorescence yield; g(t), function describing the kinetics of Q3

A reoxidation; g(t)=K1exp(3t/d1)+K2exp(3t/d2)+K3exp(3t/d3) ; HF1, Fm induced by ¢rst ST excitation in dark-adapted cells ; HF2, Fm induced by ST £ash applied following MT £ash;HFM, Fm induced by MT excitation; i(t), excitation intensity at time t in FRR protocol; I(t), cumulative excitation energy in FRRprotocol; LED, light-emitting diode; LF, Fm induced by ST excitation; MT, multiple turnover; p, extent of energy transfer between PS IIreaction centers; PAM, pulse amplitude modulation £uorometry; PpP, pump-and-probe £uorometry; PQ, plastoquinone pool; PS II,Photosystem II; qp(t), photochemical quenching at time t during the FRR protocol; QA, the primary quinone electron acceptor in PS II;QB, the secondary quinone electron acceptor in PS II; cPS II, functional (i.e., the photochemically e¡ective) cross section of PS II; RC II,reaction center of PS II; ST, single turnover; ST1, ST2, ST £ashes applied before and after the MT £ash, respectively

* Corresponding author. Fax: +1 732-932-8578; E-mail : falko@ahab.rutgers.edu1 Present address: Institute of Microbiology, MBU, AVCí R, 379 81 Tr­ebon­, Czech Republic.

BBABIO 44679 29-9-98

Biochimica et Biophysica Acta 1367 (1998) 88^106

been proposed that relate variations in £uorescenceyields to: (1) the quantum e¤ciency of photochem-istry in PS II reaction centers [3^5]; (2) the numberof primary and secondary electron-acceptors [6,7];(3) the concentration of P�680, donor side turnover,and the state of the water-splitting system [8^10];(4) the functional (or e¡ective) absorption cross-sec-tion of PS II [11^16]; (5) energy transfer betweenreaction centers [17,18]; and (6) the kinetics of elec-tron transport on the acceptor side of PS II [19^24].These parameters provide information on the funda-mental biophysical properties of photosynthetic en-ergy conversion [25,26], the e¡ects of environmentalstresses on PS II function in vivo [27], photosyntheticelectron transport under ambient irradiance [28], andthe molecular relationships between the structure andfunction of the photosynthetic apparatus [29].

Experimentally, evaluating speci¢c photosyntheticparameters partly depends upon the particular £uo-rometric method, instrument, and measurement pro-tocol used. Consequently, comparing published re-sults is complicated by di¡erences in experimentaltechniques, and has led to contention about the val-idity and weaknesses of speci¢c ones [14,18,30^34].Confusion on the quantitative interpretation of var-iable £uorescence yields stems largely from di¡eren-ces between ST and MT excitation protocols [35^37].

The oldest, and easiest, technique for measuringchanges in £uorescence yields is based on the analysisof a £uorescence transient induced in a dark-adaptedsample by a short, rapid exposure to continuous light[38,39]. This so-called `£uorescence induction techni-que' is beguilingly simple; however, because the rateof excitation delivery to PS II reaction centers (i.e.,the rate of photochemistry) in a dark-adapted state isgenerally lower than the initial rate of Q3

A oxidation,the observed transient saturates only after reductionof the plastoquinone pool [6]. As the stoichiometryof PQ to PS II is invariably s 1.0 in vivo, the ki-netics of the changes in £uorescence yields are com-plicated by multiple turnovers of PS II. This compli-cation can be avoided by applying 3-(3,4-dichlorophenyl)-1,1-dimethylurea (DCMU) or simi-lar herbicides that prevent reoxidation of Q3

A onthe relevant time scales of measurements. Use ofsuch inhibitors, however, is not practical in in situapplications.

Subsequently, non-destructive ST and MT £uores-

cence-based techniques have been developed that ex-pose samples to one or more £ashes of saturatingactinic light, and then follow changes in £uorescenceas the electron acceptors in PS II are reduced. Onewidely used technique is the so-called `light doubling'[40], or `pulse amplitude modulated' (PAM method),that uses an MT actinic pulse to induce the maxi-mum £uorescence level, as advocated by Shreiber[41]. An alternative to the MT approach is basedon `pump-and-probe' (PpP), or `£ash photolysis'techniques ([24,42^48]; see reviews in Refs.[39,48,49]). The PpP method compares £uorescenceyields for weak probe £ashes before and after an STactinic `pump' £ash. Varying the intensity of the ac-tinic £ash allows the functional absorption cross-sec-tion of PS II to be derived [25] and, by varying thedelay between actinic- and probe-£ashes, the kineticsof electron transfer on the acceptor side of PS II alsocan be evaluated [23,24,50,51].

Both PAM and PpP techniques are superior to the£uorescence induction method because they permitmore robust biophysical analyses; however, the re-quired experimental protocols last up to several mi-nutes, making it di¤cult to follow dynamic changesin the kinetics of electron transport and cPS II occur-ring in time-scales of milliseconds to minutes. In ad-dition, both techniques use excitation protocols thatpotentially change the redox state of electron carriersbetween PS II and PSI, and/or the level of nonpho-tochemical quenching within the photosynthetic ap-paratus. Consequently, the measured variable £uo-rescence critically depends upon the measurementprotocol. Speci¢cally, it was suggested that thePpP technique can induce so-called `donor-sidequenching' [8,10,43,52], whereas the PAM techniquemay mediate a di¡erent type of quenching resultingfrom excitation-induced changes in the redox state ofthe PQ pool [36,37,51].

Here we describe a novel methodology, called `fastrepetition rate' (FRR) £uorescence, that measures asuite of photosynthetic parameters, and examine howexperimental protocols a¡ect the measured quantumyields. Using this FRR approach, we can independ-ently manipulate photochemical quenching, non-pho-tochemical quenching, and the redox state of theacceptor and donor side of PS II, thereby assessingthe contribution of each process to the variable £uo-rescence yield. Our goal is to extend the application

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106 89

of £uorescence-based measurements of photosyn-thetic performance, while clarifying pragmatic issuesof the applying experimental data to interpret thebiophysical phenomena. In another paper, we exam-ine the processes responsible for the variation in thequantum yields of £uorescence [53].

2. Materials and methods

2.1. Fast repetition rate £uorometry: theory

The FRR technique measures £uorescence transi-ents induced by a series of brief subsaturating exci-tation pulses, or `£ashlets,' where the intensity, dura-tion, and interval between them is independentlycontrolled. This £exibility permits selective manipu-lation of QA and PQ reduction and allows changes in£uorescence yield to be separately attributed to thereduction states of these two acceptors. The FRRtechnique allows cPS II and the energy transfer be-tween PS II reaction centers to be calculated fromthe kinetics of £uorescence transients within a singlephotochemical turnover of PS II.

The £uorescence yield measured at time t duringan FRR excitation protocol, f(t), is de¢ned by theFo (minimal) and Fm (maximal) components of the£uorescence yield, and by the fraction of closed PS IIreaction centers, C(t). If the ensemble of PS II re-action centers share excitation energy [17,18,26,54],the observed £uorescence yield also depends on theextent of energy transfer, p :

f �t� � Fo � �Fm3Fo� C�t� 13p13C�t�p

� �: �1�

Eq. 1 also can be expressed in terms of the fractionof open RC IIs, q(t)=13C(t), but at the cost of amore complicated formalism [26].

C(t) is controlled by the rate of primary photo-chemistry, which is proportional to the product ofincident excitation energy (I), the functional absorp-tion cross-section (cPS II), and the rate of Q3

A reoxi-dation. Changes in C(t) due to primary photochem-istry can be described as:

DC�t�DI� cPS II

13C�t�13C�t�p: �2�

Ignoring Q3A reoxidation (or assuming a single turn-

over character of the excitation), Eq. 2 simpli¢es to:

dC�t�dt� cPS II

dIdt

13C�t�13C�t�p � cPS II i�t� 13C�t�

13C�t�p �3�

where i(t) is the excitation intensity. Integrating Eq.3 allows the expression of C(t) as:

C�t� �Z t

0cPS II i�e� 13C�e�

13C�e�pde; �4�

where e is the integration variable, and C(e= 0) isthe fraction of PS II reaction centers closed beforethe FRR excitation. Substituting Eq. 4 into Eq. 1allows calculation of the £uorescence transient,f(t), when the rate of excitation delivery to RC II(i.e., the product of i(t) and cPS II) greatly exceedsthe rate of Q3

A reoxidation. Such conditions exist atvery high excitation intensities, or when Q3

A reoxida-tion is inhibited by a herbicide, such as DCMU. Inthe more general case, where Q3

A reoxidation is sig-ni¢cant, Eq. 4 can be modi¢ed to account for thiscompeting process:

C�t� �Z t

0cPS II i�e� 13C�e�

13C�e�pg�t3e�de �5�

where g(t3e) describes the rate of Q3A reoxidation,

expressed as a sum of exponential components:

g�t3e� � g�vt� � K1exp�3vt=d1�

�K2exp�3vt=d2� � K3exp�3vt=d3�: �6�

Although there are no analytical solutions for Eqs.4 and 5, the photosynthetic parameters (Fo, Fm,cPS II, p, as well as the kinetic constants in Eq. 6)can be calculated by numerically ¢tting the measured£uorescence transient to a discrete form of Eq. 1:

f n � Fo � �Fm3Fo�Cn13p

13Cnp�7�

where fn is the £uorescence yield measured at the nth£ashlet, and Cn is the fraction of RC II closed at thenth £ashlet. Cn can be approximated recursively as:

Cn � Cn31

Xm

k�1

An;k � IncPS II

13 Cn31

Xm

k�1

An;k

!

13p Cn31

Xm

k�1

An;k

!�8�

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^10690

where In is the excitation energy delivered at the nth£ashlet, m is the number of exponential componentsin Eq. 6, and An;k is determined by the kinetics of Q3

Areoxidation:

An;k � �An31;k � Cn31Kk=cPS II�exp�3vt=dk�: �9�In practice, we assume that the kinetics of Q3

A reox-idation (i.e., the amplitudes, Kk, and the time con-stants, dk, in Eq. 6 and Eq. 9) are constant during theexcitation protocol. Q3

A reoxidation can be calculatedmore rigorously by performing multiple numericalanalyses for subsets of experimental data by individ-ually evaluating Kk and dk while holding Fo, Fm, p,and cPS II constant.

Ignoring Q3A reoxidation, Eq. 1 and Eq. 4 can be

combined as follows:

C�t��cPS II

Z t

0i�e�Fm3f �e�

Fm3Fode�cPS II

Z t

0�i�e�qp�e��de;

�10�

where qp(t)=[Fm3f(t)]/[Fm3Fo] is the photochemi-cal quenching. In absence of energy transfer (i.e.,when p=0) qp=q=13C. The functional absorptioncross-section for the photochemical target can there-fore be calculated as:

cPS II �Z r

0�i�e�qp�e��de

� �31 �11�

or, in discrete form:

cPS II �XN

i�0

�Inqp;n�" #

31 �12�

where N is the £ashlet number at which the observed£uorescence signal saturates at the maximum level,Fm, and qp;n is the photochemical quenching meas-ured during the nth £ashlet. Eq. 12 gives an initialestimate of cPS II for iteratively calculated photosyn-thetic parameters using Eqs. 7^9.cPS II describes the maximal e¤ciency of light uti-

lization for photochemistry in PS II in units of Aî 2/quanta:

cPS II � DCDI

MC�0 �13�

cPS II itself is controlled by light absorption and thequantum yield of photochemistry in the reaction

centers:

cPS II � DEa

DIDCDEa

MC�0 � aPS IIxmaxPS II �14�

where Ea is the absorbed excitation energy, xmaxPS II is

the maximum quantum yield of PS II photochemis-try, and aPS II is the optical absorption cross-sectionof PS II. Assuming that xmax

PS II can be assessed fromvariable £uorescence and from the extent of energytransfer between photosynthetic units [5,55], the op-tical absorption cross-section can be calculated fromFRR measurements of cPS II, Fo, Fm, and p.

Under steady-state conditions (C=13q=const) therate of excitation trapping is balanced by the rate ofQ3

A reoxidation. From Eq. 3 it follows that underconstant ambient light:

cPS IIioq�io�

�13p� � pq�io� � 13q�io�� � 1dQA�io� �15�

where io is the incident irradiance, dQA�io� is the ir-radiance-dependent, average time-constant of Q3

A re-oxidation, and q(io) is the steady-state fraction ofopen RC II. Rearranging Eq. 15 we calculate q(io)as follows:

q�io� � c0PS II�io�io

c0PS II�io�io � 1dQA

�16�

where

c0PS II�io� � cPS II

�13p� � pq�io� �17�

re£ects an increase in the functional absorptioncross-section at intermediate irradiance levels dueto excitation energy transfer between PS II reactioncenters.

Both cPS II and aPS II are wavelength-dependent;using excitation light of controlled spectral quality,the FRR technique allows measurements of the spec-trally resolved functional absorption cross-section,cPS II(V). In this report, we limit our discussion tomeasurements of cPS II at 475 nm, with 70 nm half-bandwidth.

2.2. Fast repetition rate £uorometry:instrumentation

The FRR instrument uses a bank of blue and green

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106 91

light emitting diodes (a combination of NLPB500,450 nm peak emission, and NSPE500, 500 nm peakemission) from Nichia, Japan, as an excitationsource. To uniformly illuminate the sample chamber(a T-76 scattering cell from NSG Precision Cells,USA), three stacks of LEDs, totaling 108, are ar-ranged as a cylinder of 64 mm diameter. Residuallong-wavelength radiation from the LEDs is ¢lteredwith a cylindrical, 3-mm-thick glass ¢lter (BG39from Shott, Germany). The LEDs are operated at300 mA pulsed current, and each produces W8mW of peak optical power. To avoid thermal dam-age to the LEDs, the duty cycle of excitation £ashesis limited to 6 5%. Peak optical power within thesample chamber is W0.7 W/cm2, corresponding toa £ux of 1.66U1022 quanta m32 s31 (i.e., 27 500Wmol quanta m32 s31).

Emitted £uorescence is collected from the bottom

of the sample chamber through a pair of interference¢lters (Corion S10-680-R) and a long pass ¢lter(Shott RG665). The £uorescence signal is detectedusing a Hamamatsu R2066 photomultiplier operatedat 300^600 V. As a reference signal, a small portionof the excitation light is detected by a HamamatsuS1722 PIN photodiode. Both the £uorescence andthe reference signals are digitized at a 16 MHz sam-pling rate, and transferred to a notebook computer.The digitized reference signal (i), and the ratio of the£uorescence to reference signal (f) then are analyzedwith custom software in the context of Eqs. 7^9.

2.3. Fast repetition rate £uorometry:excitation protocols

The £exibility of the FRR protocol, which can lastfrom tens of microseconds to seconds, permits both

Fig. 1. Representative £uorescence transients measured by the FRR method. Phases I and V (ST £ashes) are performed as a sequenceof 80 to 120 £ashlets of 0.125 to 1.0 Ws duration, at 0.5- to 2.0-Ws intervals. Phases II, IV, and VI (relaxation protocol) consist of 40to 80 £ashlets at intervals varying exponentially from 50 Ws to 50 ms. Phase III (MT £ash) consists of up to 4000 £ashlets, at 20- to200-Ws time intervals. The length of Phase IV is varied from 1 Ws to 10 s by changing the number of £ashlets and time intervals be-tween them. The £uorescence transients in Phase I and V are controlled by the Fo and Fm components of the £uorescence yield, thefunctional absorption cross-section (cPS II), and by the extent of energy transfer between PS II reaction centers, p. Fluorescence transi-ents in Phases II and VI are controlled primarily by the kinetics of Q3

A reoxidation, while the £uorescence transient in Phase IV iscontrolled primarily by the kinetics of PQ pool reoxidation. Phase III measures changes in £uorescence yield under multiple turnoverconditions, and Phases V and VI quantify the e¡ects of the earlier multiple turnover excitations on the photosynthetic parameters.

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^10692

ST and MT excitation sequences to be executed inany combination and arbitrary delay time. Duringthe ST protocol, high excitation energies are usedto reduce QA within a single photochemical turnover.Operationally, the ST £ash consists of a series of80^120 £ashlets, each of 0.125 to 1.0 Ws duration,at 0.5 to 2.0 Ws intervals, and a pulse power ofV0.03 mol quanta m32 s31 (Fig. 1, Phase I). Thecumulative excitation energy during an ST £ash isselected so that 3 to 4 quanta per RC II are ab-sorbed. Inevitably, a portion of PS II reaction centerswill become reoxidized and potentially can be re-re-duced during the £ash, thus deviating from a purelysingle-turnover excitation. We discuss the extent ofthis deviation below.

During the MT protocol, excitation energy is de-livered over a longer time, allowing multiple QA ox-idation^reduction events, and reducing the PQ poolwithin 60 to 500 ms. Operationally, the MT protocolconsists of a sequence of up to 4000 £ashlets each of0.125 to 32 Ws duration, delivered at intervals of 20 Wsto 2 ms (Fig. 1, Phase III).

In addition to the ST and MT excitation £ashes,the FRR protocol includes relaxation sequences toevaluate the kinetics of Q3

A reoxidation (Eq. 6).These sequences consist of 40 to 80 low-energy £ash-lets at 20-Ws to 50-ms intervals, and are deliveredafter either an ST or MT £ash (Fig. 1, Phase IIand IV). Usually, we follow the MT £ash by a sec-ond ST/relaxation sequence at varying intervals (Fig.1, Phase V and VI).

2.4. Experimental materials

All the results discussed in this paper are from theunicellular chlorophyte, Chlorella pyrenoidosa, andare representative of the FRR methodology appliedto a wide variety of cultured, unicellular algae (e.g.,Dunaliella tertiolecta, Chlamydomonas reinhardtii,and Skeletonema costatum), as well as in natural phy-toplankton communities, and intact leaves fromhigher plants and blades of marine macrophytes.Chlorella pyrenoidosa was selected because it canbe easily grown with minimal algal culturing facili-ties, making it the preferred organism to test andreproduce the methodology and results presentedhere.

2.5. Oxygen evolution

Oxygen £ash yields were measured using a bareplatinum oxygen-rate electrode as described in Ref.[37]. ST £ashes for oxygen measurements were gen-erated using a xenon £ashlamp (model L21 fromHamamatsu). The £ashlamp was operated in aFRR mode by applying a triggering HV pulse fol-lowed by a series of high current pulses at 1.5- to 10-Ws intervals, resulting in a train of 4 to 80 £ashletsdelivering a total cumulative energy of 3 quanta/RCII.

3. Results

3.1. Single turnover £ashes

To investigate the relationship between excitationenergy and £uorescence responses, we applied an ST£ash consisting of 120 £ashlets of 0.2 Ws to 0.8 Wsduration at 1-Ws intervals (Fig. 2A). With increasingexcitation energies, the £uorescence yield saturatesfaster, reaching the Fm level within 60 Ws. All thepro¢les displayed identical saturation kinetics as afunction of cumulative excitation energy (Fig. 2B).Assuming that the steady-state £uorescence yield iscontrolled primarily by the ratio of the rates of QA

reduction/reoxidation, these results suggest that,under our experimental conditions, the averagetime constant of Q3

A reoxidation is much longerthan the length of the £ash. Hence, only a smallfraction of Q3

A is reoxidized during the £ash; thisfraction is independent of excitation energy. The ob-served £uorescence transients deviate from a cumu-lative one-hit Poisson distribution [13] (Fig. 2B, in-set), suggesting energy transfer between PS II units.The calculated values of Fo, Fm, cPS II and p areindependent of excitation energy between 1.6 to 4quanta/RC II/ST (Table 1). Although the £uores-cence transients observed within 120 ms £ash neverreach the theoretical maximum level (Fig. 2A), thecalculated asymptotic Fm signal is independent on theexcitation energy within this range.

To further examine the e¡ect of excitation energyon the saturation character of the £uorescence tran-sients, we applied a series of 120 £ashlets of 0.8 Wsduration, but altered the intervals between them

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106 93

from 1 Ws to 50 Ws (Fig. 2C). The rate of £uorescencesaturation normalized to cumulative excitation en-ergy is constant during the initial 100^200 Ws (Fig.

2D). This rate declines at longer £ash lengths withdecreasing excitation energy, indicating an increasinge¡ect of Q3

A reoxidation.

Fig. 2. E¡ect of excitation energy on £uorescence saturation pro¢les observed during an ST protocol consisting of 120 £ashlets. (A)Excitation energy was controlled by varying the duration of the £ashlets from 0.2 Ws to 0.8 Ws, while maintaining the time interval be-tween them, at 1 Ws. When plotted as a function of excitation energy (B), all the £uorescence transients displayed a similar saturationcharacter during the 120-Ws time window. Inset: an example of a £uorescence transient as a function of the log cumulative excitationenergy. The solid line is experimental data, the dotted line is the ¢t of the data to a cumulative one-hit Poisson model assuming noenergy transfer between PS II reaction centers. By neglecting energy transfer, the least squares minimization procedure forces the cal-culated cPS II to decrease relative its correct value in order to minimize the error between the calculated and the experimental curve.(C) Excitation energy was controlled by varying the time interval between £ashlets from 1 Ws to 50 Ws, while maintaining the length0.8 Ws. The resulting £uorescence transients showed slower saturation rates as a function of time (C), and cumulative excitation energy(D), indicating that at longer time windows the £uorescence transient is controlled by the excitation energies as well as by the rates ofQ3

A reoxidation.

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^10694

The excitation energies selected for ST protocolsare a compromise between a desire to fully reduceQA within a single photochemical turnover, and thedesire to minimize potential non-photochemical £uo-rescence quenching in PS II reaction centers causedeither by carotenoid triplet formation [56,57] or P�680accumulation [8,9,52]. The latter two phenomenamay reduce £uorescence yields by 20^35% duringST £ashes compared with MT £ashes [35]. We did

not observe marked changes in Fm when the excita-tion intensity was decreased by a factor of three (Ta-ble 1), nor in the initial rise of £uorescence yieldwhen the excitation intensity was decreased by a fac-tor of 50 (Fig. 2B,D). However, Fv/Fm measured withan ST £ash was approx. 0.65 in physiologicallyhealthy algae, which is considerably lower than thatof 0.75 which is sometimes observed using an MTprotocol (e.g., Ref. [58]).

Table 1Calculated values of Fo, Fm, cPS II, and p for single-turnover £ashes with di¡erent excitation energies (Fig. 2A,B) using Eq. 8, and as-suming negligible Q3

A reoxidation during the £ash

Flashlet length (Ws) Excitation energy (quanta/RC II/ST) Fo Fm cPS II (Aî 2/q) p

0.8 4.1 1.00 2.72 280 0.440.6 3.0 1.00 2.72 290 0.430.5 1.9 1.00 2.72 290 0.440.4 1.6 1.00 2.72 300 0.430.3 1.2 1.00 2.59 310 0.380.2 0.6 1.00 ^ ^ ^

The lower limit of excitation energies to reliably calculate photosynthetic parameters is about 1.6 quanta/RC II ST under the condi-tions of negligible Q3

A reoxidation. Fm values shown here are the calculated, asymptotic maximum £uorescence yield, not the £uores-cence yields attained at the end of the ST £ash.

Fig. 3. FRR £uorescence transients observed during an ST £ash followed by a relaxation protocol. The ST protocol consisted of 100£ashlets of 0.036 q/RC II with an interval of 1 Ws. The relaxation protocol consisted of 40 £ashlets at 20-Ws intervals, where the exci-tation energy varied from 0.002 to 0.022 quanta/RC II. (b) Experimental data, (9) theoretical ¢t with parameters calculated as shownin Table 2. (- - -) Cumulative excitation energy.

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106 95

3.2. FRR relaxation protocol

Following the transient reduction of QA by an ST£ash, the decay of £uorescence is controlled by Q3

Areoxidation kinetics [22], but is confounded by en-ergy transfer between PS II reaction centers [59].The kinetics of Q3

A reoxidation can be described bya multi-exponential decay model (e.g., Eq. 6), andcan be assessed with the FRR £uorometer usingthe described relaxation protocol. Energetically, theexcitation energy used in this protocol has an actinice¡ect that repopulates a fraction of the reoxidizedQA. These re-excited RC IIs display reoxidation ki-netics that deviate from a single-turnover character.To investigate the e¡ect of the relaxation protocol onthe assessment of Q3

A reoxidation kinetics, we ana-lyzed the £uorescence responses acquired at varyingexcitation energies; Fig. 3 demonstrates the actinice¡ect of the measurement protocol. These artifactscan be minimized by using excitation protocols thatproduce the least actinic e¡ect, and by numericalanalysis that accounts for the inevitable re-reductionof a fraction of centers by the probe £ashlets (e.g.,Eqs. 8 and 9).

The kinetics of Q3A reoxidation calculated with

such a correction (Table 2) are almost independentof the excitation energy. Deviation in the calculatedparameters at the highest level of excitation energymost likely re£ects changes in the population of re-duced secondary electron- carriers (i.e., QB and PQpool).

The cumulative energy applied during the STportion of the excitation protocol was about 3.6quanta/RC II, resulting in £uorescence transientfrom 1 to 2.84, and about 0.96 level of QA reductionat the end of the ST £ash. The theoretical Fm signal

calculated using Eqs. 7^9, however, was around3.0.

3.3. Excitation energy transfer

Energy transfer between PS II reaction centers,represented by the p parameter in Eq. 1 and Eq. 5,a¡ects both the rate of £uorescence saturation duringan ST £ash and £uorescence decay during relaxationprotocol. As p increases, the rate at which £uores-cence yields decay following the ST £ash exceeds therate of Q3

A reoxidation. Consequently, neglecting pwhen analyzing the Q3

A reoxidation underestimatesthe kinetic time constants, particularly the fast andmedium components (last row in Table 2). Applyingan arbitrary value for p may alleviate this problem[59]; however, our measurements indicate that pvaries widely, and strongly depends upon the phys-iological condition of the cells [60]. Therefore, assess-ing the rates of Q3

A reoxidation from £uorescenceyield decay-curves requires calculating p on a case-by-case basis.

3.4. How `single turnover' are ST £ashes?

By de¢nition, an ST £ash induces a single photo-chemical event. This requirement can be evaluated bycalculating the extent of Q3

A reoxidation by QB dur-ing the ST £ash:

et �XN

n�0

Cn�K1exp�3vtn=d1� � K2exp�3vtn=d2�

�K3exp�3vtn=d3�� �18�

where et is a measure of the electron equivalents

Table 2Fluorescence parameters and time constants of Q3

A reoxidation for di¡erent excitation energy of the relaxation protocol (Fig. 3A,B),analyzed with the triple-exponential model of Q3

A reoxidation, according to Eqs. 8^10

Flashlet energy q/RC II Fo Fm cPS II (Aî 2/q) p K1 d (Ws) K2 d2 (Ws) K3 d3 (ms)

0.002 1.00 3.01 300 0.48 0.44 240 0.44 1240 0.12 26.90.004 1.00 2.99 320 0.49 0.41 250 0.44 1300 0.15 26.50.006 1.00 2.99 330 0.48 0.42 260 0.43 1360 0.15 26.00.013 1.00 2.96 310 0.45 0.40 270 0.42 1230 0.18 20.60.022 1.00 2.96 310 0.45 0.33 360 0.48 1280 0.21 12.40.002 1.00 3.00 350 0.00 0.530 180 0.380 830 0.090 29.2

The results in the last row are an example of calculations of the kinetics of Q3A reoxidation, neglecting energy transfer (p = 0).

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^10696

Fig. 5. Oscillation pattern of £ash oxygen yields in response to a series of £ashes with cumulative energy of 3 quanta/RC II andlength of 6 to 640 Ws, applied at 2-s time intervals. Each £ash consisted of a train of 4 to 80 £ashlets. The oscillatory pattern of oxy-gen evolution is preserved during the ¢rst four £ashes of length varying from 6 to 160 Ws, although its amplitude dampens with in-creasing £ash length due to increasing probability of double-electron turnovers. The oxygen evolved during the second £ash, and inthe steady state (s 24 £ashes) increases by about 8% as the length of the £ash increases from 6 Ws to 80 Ws, indicating 6 10% in-crease in the probability of double electron turnover during an 80-Ws-long £ash.

Fig. 4. Fluorescence saturation pro¢les and initial portion of relaxation protocol. The ST £ashes consisted of a train of £ashlets, ap-plied at times varying from 1.6 Ws to 12 Ws (b). The relaxation kinetics were measured with £ashlets applied at time intervals variedexponentially from 50 Ws to 800 ms. (- - -) The calculated electron transfer from Q3

A to secondary electron acceptors. The continuousline represents a theoretical ¢t to the experimental data, using Eqs. 8 and 9.

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106 97

resulting from Q3A reoxidation occurring during N

£ashlets, vtn is the interval between £ashlets, andCn is calculated by Eq. 8. To quantify et we madea series of ST measurements with time intervals be-tween £ashlets varying from 1.6 Ws to 12 Ws, followedby a relaxation protocol (Fig. 4). From Eqs. 8 and 9,we calculate that up to 96% of PS II reaction centerswere reduced within 60 Ws at the highest excitationintensity, 6 10% of the reaction centers were reoxi-dized, and 6 7% were re-reduced by subsequent ex-citation during the ST £ash. We conclude that thedelivery of 3 to 4 excitations per RC II within 60^100 Ws is e¡ectively an ST £ash. Under such condi-tions, 6 10% of the PS II reaction centers undergomultiple photochemical turnovers. As the time be-tween £ashlets was increased from 1.6 to 12 Ws, thenumber of multiple turnovers increased by the samefactor.

To independently establish the single turnovercharacter of ST £ashes, we measured oxygen £ashyields using a train of £ashlets with a total durationof 6 to 640 Ws; these conditions were achieved usinga sequence of 4 to 64 £ashlets at 1.5- to 10-Ws inter-vals, with constant cumulative excitation energy of 3quanta/RC II. A four-£ash periodicity in oxygen and£uorescence was observed, which dampened after 12£ashes (Fig. 5). Changing the length of the excitationsequence from 6 to 160 Ws had little e¡ect on theinitial (¢rst to fourth £ash) oscillatory pattern forO2 evolution (Fig. 5). The oxygen-£ash yield on thesecond £ash and in the steady-state (s 24 £ashes)increased by about 8% as the length of the £asheschanged from 6 Ws to 80 Ws. We conclude that thenumber of double photochemical turnovers in an 80-Ws-long ST £ash increases by 6 10% when the £ashlength rises from 6 Ws to 80 Ws.

3.5. Dependence of £uorescence yield on £ash numberduring the ST protocol

If ST £ashes are applied at intervals 6 4 s, the£uorescence signal observed in a dark-adapted sam-ple displays a high £uorescence yield during the ¢rst£ash (HF1), followed by a weak, four-period oscil-lation with a lower (LF) £uorescence yield (Fig. 6)[42,61^63]. The oscillation pattern of £uorescence isgenerally antiparallel to that of oxygen evolution[42,64], and is absent under weak background light

or in samples immediately taken from illumination tothe dark. The oscillatory pattern is independent ofthe ST £ash length in the 6^160 Ws range, but damp-ens with longer £ash durations due to multiple turn-overs of PS II that randomize S-states. These resultspreviously were interpreted as donor-side modulationof variable £uorescence [42,65]. The pattern of Fm

changes in a given £ash is preserved in the Fo signalon the subsequent £ash (Fig. 6A), suggesting thatthis modulation is maintained over the lifetime ofthe S state (seconds) [21,24,42], rather than the life-time of oxidized P�680 (sub-milliseconds) [66]. Themodulation of the £uorescence yield by the donorside of PS II shown in Fig. 6A should not be con-fused with `donor-side quenching' of the Fm signal[10]; the modulation reported here is preserved sev-eral seconds after an ST £ash. On this time scale,potential quenchers in PS II reaction centers (e.g.,P�680 and/or carotenoid triplets) have long dissipated[57,66].

When the time delay between £ashes is increasedto 10 s, the oscillations in £uorescence yield disap-peared; however, the ¢rst £ash still had a 15% to20% higher yield than that of the second and subse-quent £ashes (Fig. 6B, and inset). The di¡erence be-tween the ¢rst and subsequent ST £ashes is inde-pendent of the length of the £ash in the 80^1000 Wsrange. When the time delay between £ashes is s 120s, only the HF £uorescence yield is observed. Whenthe sample is dark-adapted for less than 10 s, orexposed to a very weak background illumination of0.1^0.5 WE m32 s31, only the LF £uorescence yield isobserved.

As the duration of the ST £ash is extended beyondapprox. 200 Ws, the £uorescence yields declineslightly from their initial maximum. Furthermore,there is a transient spike at the onset of the relaxa-tion sequence, indicating the development of £uores-cence quenching during the excitation protocol. Theextent of this quenching (3^5%), however, is muchsmaller than the di¡erence between the HF and LFlevel.

3.6. MT £ash protocol

Fluorescence transients observed during an MT£ash display complicated patterns of £uorescenceyield in response to progressive excitation (Fig.

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^10698

7A). The multiphasic character of the £uorescencesignal re£ects sequential ¢lling and emptying of sev-eral electron pools (QA, QB, PQ) with di¡erent re-duction-oxidation rates. The expression of thesephases strongly depends on the experimental proto-col [67^69].

As with ST £ashes, the £uorescence transients ob-

served at the ¢rst and the second MT £ash in a dark-adapted sample are very di¡erent. Fig. 7B showsrepresentative £uorescence pro¢les for three pairsof MT £ashes, each composed of 3000 £ashlets, ap-plied at 10-s intervals, with average excitation ener-gies of 4000, 185, and 40 quanta/RC II/s. The ¢rstMT £ash displays a higher maximal £uorescence

Fig. 6. ST £uorescence transient as a function of the £ash number. (A) Period four oscillation of the Fm, Fo, and intermediate (Fdel)£uorescence yield observed in a series of 100-Ws-long ST £ashes applied at 2-s intervals. The Fdel £uorescence signal was measured 1to 16 ms following the ST £ash. High Fo £uorescence at given £ash correlates with the high Fm £uorescence at the preceding £ash.(B) Dependence of £ash number of the £uorescence transient observed in a dark-adapted sample (tdark s 120 s, thin line) and shortdark-adapted sample (tdark 6 10 s, thick line). The long £ashes consist of 640 £ashlets of 0.8 Ws duration and 1-Ws intervals. Long£ashes are separated by a relaxation sequence consisting of 120 £ashlets applied at times exponentially increasing from 50 Ws to 30ms, over a period of 10 s. The short £ashes (80 £ashlets) are separated by a relaxation sequences consisting of 40 relaxation £ashletsover a period of 2 s. Inset: the di¡erence between the ¢rst and second ST £ash applied on dark-adapted sample (tdark s 120 s); high£uorescence yield (HF1) is observed during the ¢rst £ash (9), and a low £uorescence yield (LF) is observed during the second andconsecutive £ashes (- - -).

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106 99

yield, with a slowly saturating phase and a lowerinitial slope compared with subsequent ones. Whenthe delay between £ashes is longer than 2 min, the¢rst £ash response is preserved with each subsequent£ash. When measured under weak background illu-mination or after 6 10 s of dark adaptation, £uores-cence yields resembled the second-£ash behavior inFig. 7B.

To further examine £uorescence responses to STand MT £ashes, we exposed cells to an ST £ash(ST1) followed by an MT £ash, and ¢nally to a sec-ond ST £ash (ST2), each separated by a relaxation

sequence (Fig. 8A). The maximum £uorescence yield,HF2, measured during the ST2 £ash depended onthe interval between MT and ST2. If the ST2 £ashwas applied within 10 Ws to 5 ms following the MT£ash, the £uorescence signal remained at a level4^5% lower than the maximum yield, HFM(Fig. 8B). As the time interval between MT andST2 increased to 16 s, the HF2 decreased, with atime constant of approx. 4.2 s. The functional ab-sorption cross-section measured 160 ms followingMTF was about 40% lower than that measured dur-ing ST1. As the interval between MTF and ST2 in-

Fig. 7. Fluorescence response to multiple turnover (MT) £ashes. (a) Fluorescence transients in MT £ashes performed with 3000 £ash-lets of 0.8 Ws duration and intervals varying from 20 to 2000 Ws. Very high variable £uorescence (Fv/Fm = 0.75) is observed at excita-tion energies as low as 200 quanta/RC II/s. (B) Fluorescence response to a pair of MT £ashes applied at 10 s delay measured in adark-adapted sample.

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106100

Fig. 8. (A) Fluorescence transients observed during a sequence of ST1^MT^ST2 £ashes. ST £ashes consist of 80 £ashlets of 0.8 Ws du-ration at 1.5-Ws time interval, MT £ash consists of 3200 £ashlets of 0.8 Ws duration or length at 20-Ws intervals (every 10th £ashlet inthe MT £ash is shown). The relaxation sequences following the ST1 and ST2 consist of 40 £ashlets at intervals varying exponentiallyfrom 50 Ws to 800 ms, over 4 s. Following an MT £ash, the time interval between £ashlets in relaxation sequence was varied from125 Ws to 2.4 s. The ST2 £ash was applied 10 Ws following an MT £ash, at 20 ms delay, and 10 s delay. Inset: schematic representa-tion of excitation protocol in the ST £ash, relaxation sequence, and MT £ash. The Fm signal observed during ST1 £ash is minimal(LF) when measured on a short, dark-adapted sample (tdark 6 10 s), or in the presence of low-level background irradiance (continuousline). It increases to the HF1 level when measured on a long dark-adapted sample (tdark s 120 s, dashed line). The Fm signal observedduring the ST2 £ash (HF2) approaches the level measured during a MT £ash when applied immediately following a MT £ash. The ki-netics of Fm decrease following the MT £ash (tFH2 = 4.2 s) is similar to the kinetics of the slow phase of changes in £uorescence yieldfollowing an MT £ash (tMT = 2.6 s). (B) Changes in Fm £uorescence yields and functional absorption cross-section as a function of thedelay between an MT and an ST2 £ash.

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106 101

creased to 16 s, cPS II reached the pre-MTF value,with a time constant of V1 s (Fig. 8B).

The results from the ST1^MT^ST2 protocol illus-trate the di¡erences in £uorescence yields for ST andMT measurements which are the basis of the contro-versy in interpreting £uorescence data [14,34,36,51,70]. By changing the period of dark adapta-tion, or the time between MT and ST2, we foundthat the di¡erence in Fm between MT and ST £asheswas controlled primarily by the pre-illumination his-tory, not by the intensity of excitation. The Fm signalobserved during MT £ash was up to 35% higher thanthat observed during ST £ash when measured onshort-term (tdark 6 10 s) dark-adapted samples. Fol-lowing longer period of dark adaptation (tdark s 120s), this di¡erence decreased to less than 20%. Finally,when measured shortly (10 Ws to 5 ms) following theMT £ash, the single-turnover Fm signal was only 5%lower than that observed during MT £ash.

4. Discussion

We have presented a novel method for character-izing PS II processes based on measuring £uores-cence transients in response to a series of subsaturat-ing £ashlets of controlled intensity, duration, andintervals. The key photosynthetic parameters calcu-lated with the FRR method include cPS II, Fo, Fm, p,and rates of electron transport within PS II. Themost controversial and protocol-dependent parame-ter is the maximal £uorescence level (Fm) measuredupon reduction of the ¢rst electron acceptor, QA,commonly used in £uorescence-based models of pho-tosynthetic yields or rates [15,28,34].

ST and MT £ashes di¡er in their excitation inten-sities, durations, and the cumulative excitation en-ergy required to induce the £uorescence responses.ST £ashes are generated with excitation intensitiesexceeding 50000 quanta/RC II/s in 6 100 Ws, where-as MT £ashes have excitation intensities rangingfrom 20 to 2000 quanta/RC II/s, over a period of50 ms to several seconds. Accordingly, the cumula-tive excitation-energy is 3 to 4 quanta/RC II for anST £ash, and 50 to 500 quanta/RC II for an MT£ash. It has been stated that the high excitation in-tensity of an ST £ash in£uences the measured £uo-rescence signal and prevents detection of the `correct'

£uorescence yield [10]. Our results (Fig. 8) demon-strate that the duration of the excitation protocoland total excitation energy applied, rather than ex-citation intensity, a¡ects the measured £uorescenceyield.

Di¡erences in £uorescence yields for ST and MTmeasurements are commonly ascribed to £uorescencequenching due to over-excitation of PS II reactioncenters, and resulting from accumulation of the oxi-dized intermediates on the donor side of PS II[8,9,52], or of carotenoid triplet states [56,57]. Weattributed only 4% to 5% £uorescence quenching toeither of these phenomena during extended, 640-Ws-long ST protocols (Fig. 6B). This quenching is negli-gible during the typical 80^120-Ws-long ST £ashes, asdemonstrated by the constant level of variable £uo-rescence at lower excitation energies (Fig. 2A,B). Theamplitude of this quenching phenomenon cannot ex-plain the di¡erences between the HF and LF states.We also observed a weak, four-period oscillation inthe ST £uorescence signal, which suggests that thedonor side of PS II modulates the £uorescence yieldsomewhat (Fig. 6A). This apparent control is exertedover the lifetime of the S-states, not the lifetime ofP�680 (Fig. 6C), and therefore may re£ect the accumu-lation of a positive charge near the manganese com-plex on the donor side of PS II [71,72]. It is possiblethat the di¡erence between HF1 and LF levels (Fig.6B) is due to a quenching by increased level of P�680,the reduction rate of which decreases during the sec-ond and subsequent £ashes where the S-states be-come progressively scrambled. This £uorescencequenching, however, disappears immediately follow-ing the MT £ash (HF2wHFM, Fig. 8A), where theS-states are also scrambled, indicating that the dis-tribution of the S-states has little control over theobserved quenching. The observed pattern of £uores-cence modulation also can be interpreted in terms of£uorescence quenching by S-state-dependent oxygenrelease [73]. Within these interpretations, both the Fo

and Fm signals are a¡ected (Fig. 6A), potentiallypreserving the Fv/Fm or Fv/Fo ratio.

Alternatively, changes in the rate of S-state ad-vancements may modify the lifetime of P�680 followinga double reduction of the PS II reaction center (e.g.,when reduction of P�680 requires prior advancementof S-states). The probability of double reduction dur-ing an ST £ash may reach 5% to 10% depending on

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106102

the experimental conditions, potentially resulting in along-lived, weak four-period oscillation during a se-quence of ST £ashes [62,74,75]. The transient declinein £uorescence induced by 640-Ws-long, high-inten-sity £ashes (Fig. 6B), or with an ST £ash appliedimmediately after an MT £ash (Fig. 8) might be ex-plained by double hits leading to an accumulation ofP�680. Within this model, the Fm signal would be pref-erentially quenched, causing a decrease in the Fv/Fm

ratio. The amplitude of this e¡ect (V5%), however,is negligible in comparison to the di¡erence betweenST and MT £uorescence yields (up to 35%). We con-clude that the comparatively high excitation intensityused in the ST £ash protocol cannot induce the ob-served di¡erence in £uorescence yields between STand MT £ashes. Consequently, such di¡erences in£uorescence yields are due either to changes in therate of excitation trapping by the reaction centers,and/or result from processes on the acceptor side ofPS II.

The di¡erences in Fm observed in ST and MT£ashes a¡ect assessments of the photosynthetic quan-tum yields based on calculating Fv/Fm [5], and soapparently underestimate xmax

PSII based on ST meas-urements [34]. As shown in Eq. 14, cPS II can be ex-pressed as the product of xmax

PSII and the optical ab-sorption cross-section of PS II. The FRR£uorescence technique described here can determinecPS II instantaneously (within approx. 100 Ws). Thedecrease in cPS II observed in the HF2 state (Fig.8B) suggests that xmax

PSII decreases as the £uorescenceyield increases during MT protocols. Our results sug-gest that the redox state on the acceptor side a¡ectsthe distribution of excitation energy between photo-chemistry and £uorescence. The simultaneous rise in£uorescence yield and decrease in cPS II suggests thatthe rates of radiative deactivation increase at the costof photochemistry. An antiparallel relationship be-tween the rates of oxygen evolution and slow phasesof £uorescence induction is well documented[38,76,77], and a parallel relationship between oxygen£ash yields and cPS II was observed [37]. These ob-servations are consistent with our ¢ndings.

In conclusion, the FRR methodology discussedhere allows exhaustive characterization of PS II £u-orescence under conditions of either ST or MT ex-citation. Using FRR £uorescence in conjunction withmolecular genetic approaches and other biophysical

techniques, we can examine the processes controllingvariations in the £uorescence yields and develop amodel to account for these variations [53].

Acknowledgements

We dedicate this paper to David Mauzerall fromRockefeller University for his counsel and guidancethrough the years. We thank Dr. Michael Behrenfeldfor his valuable comments on the manuscript, andKevin Wyman for his technical assistance. This re-search was supported by the U.S. Department ofEnergy under Contract No. DE-AC02-76CH0000016, and the O¤ce of Naval Researchunder Grant 97PR00617-00.

References

[1] P.G. Falkowski, Z. Kolber, Variations in chlorophyll £uo-rescence yields in phytoplankton in the world oceans, Aust.J. Plant Physiol. 22 (1995) 341^355.

[2] G.H. Krause, E. Weis, Chlorophyll £uorescence and photo-synthesis ^ the basics, Annu. Rev. Plant Physiol. Plant Mol.Biol. 42 (1991) 313^349.

[3] W.L. Butler, M. Kitajima, Energy transfer between PSII andPSI in chloroplasts, Biochim. Biophys. Acta 396 (1975) 72^85.

[4] E. Weis, J.A. Berry, Quantum e¤ciency of photosystem II inrelation to `energy' ^ quenching of chlorophyll £uorescence,Biochim. Biophys. Acta 894 (1990) 198^208.

[5] B. Genty, J.M. Briantias, N.R. Baker, The relationship be-tween the quantum yield of photosynthetic electron trans-port and quenching of chlorophyll £uorescence, Biochim.Biophys. Acta 990 (1989) 87^92.

[6] S. Malkin, B. Kok, Fluorescence induction studies in iso-lated chloroplastsI. Number of components involved in thereaction and quantum yields,, Biochim. Biophys. Acta 126(1966) 413^432.

[7] N. Murata, M. Nishimura, A. Takamiya, Fluorescence ofchlorophyll in photosynthetic systemsII. Induction of £uo-rescence in isolated spinach chloroplasts,, Biochim. Biophys.Acta 120 (1966) 23^33.

[8] W.L. Butler, On the primary nature of £uorescence yieldchanges associated with photosynthesis, Proc. Natl. Acad.Sci. U.S.A. 69 (1972) 3420^3422.

[9] A. Sonneveld, H. Rademaker, L.N.M. Duysens, Chlorophylla £uorescence as a monitor of nanosecond reduction of thephotooxidized primary donor P�680 of Photosystem II, Bio-chim. Biophys. Acta 548 (1979) 536^551.

[10] U. Schreiber, C. Neubauer, The polyphasic rise of chloro-phyll £uorescence upon onset of strong continuous illumina-

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106 103

tionII. Partial control by the photosystem II donor side andpossible ways of interpretation,, Z. Naturforsch. 42c (1987)1255^1264.

[11] C. Bonaventura, J. Myers, Fluorescence and oxygen evolu-tion from Chlorella pyrenoidosa, Biochim. Biophys. Acta 189(1969) 366^383.

[12] D.C. Mauzerall, Multiple excitation and the yield of chloro-phyll a £uorescence in photosynthetic system, Photochem.Photobiol. 28 (1978) 991^998.

[13] A.C. Ley, D.C. Mauzerall, Absolute absorption cross-sec-tions for photosystem II and the minimum quantum require-ment for photosynthesis in Chlorella vulgaris, Biochim. Bio-phys. Acta 680 (1982) 95^106.

[14] P.G. Falkowski, K. Wyman, A.C. Ley, D.C. Mauzerall, Re-lationship of steady-state photosynthesis to £uorescence ineucaryotic algae, Biochim. Biophys. Acta 849 (1986) 183^192.

[15] B. Genty, J. Harbinson, J.M. Briantais, N.R. Baker, Therelationship between non-photochemical quenching of chloro-phyll £uorescence and the rate of photosystem II photo-chemistry in leaves, Photosynth. Res. 25 (1990) 249^257.

[16] G. Samson, D. Bruce, Complementary changes in absorp-tion cross-sections of photosystems-I and photosystems-IIdue to phosphorylation and Mg2� depletion in spinach thy-lakoids, Biochim. Biophys. Acta 1232 (1995) 21^26.

[17] G. Paillotin, Capture frequency excitation and energy trans-fer between photosynthetic units in the Photosystem II,J. Theor. Biol. 58 (1976) 219^235.

[18] A.C. Ley, D.C. Mauzerall, The extent of energy transferamong Photosystem II reaction centers in Chlorella, Bio-chim. Biophys. Acta 850 (1986) 234^248.

[19] B. Bouges-Bocquet, Electron transfer between the two pho-tosystems in spinach chloroplasts, Biochim. Biophys. Acta314 (1973) 250^256.

[20] B. Diner, Dependence of the turnover and deactivation re-actions of Photosystem II on the redox state of the pool Avaried under anaerobic conditions, in: Proceedings of theThird International Congress on Photosynthesis II, Elsevier,Amsterdam, 1974, pp. 589^601.

[21] J. Bowes, A.R. Crofts, Binary oscillations in the rate ofreoxidation of the primary acceptor of Photosystem II, Bio-chim. Biophys. Acta 590 (1980) 373^384.

[22] A.R. Crofts, C.A. Wraight, The electrochemical domain ofphotosynthesis, Biochim. Biophys. Acta 726 (1983) 149^185.

[23] H.H. Robinson, A.R. Crofts, Kinetics of the oxidation-re-duction reactions of the photosystem II quinone acceptorcomplex, and the pathway for deactivation, FEBS Lett.153 (1983) 221^226.

[24] D.M. Kramer, H.R. Robinson, A.R. Crofts, A portablemulti-£ash kinetic £uorimeter for measurement of donorand acceptor reactions of photosystem 2 in leaves of intactplants under ¢eld conditions, Photosynth. Res. 26 (1990)181^193.

[25] P.G. Falkowski, Z. Kolber, Y. Fujita, E¡ect of redox stateon the dynamics of photosystem II during steady-state pho-

tosynthesis in eucaryotic algae, Biochim. Biophys. Acta 933(1988) 432^443.

[26] J. Lavergne, H.W. Trissl, Theory of £uorescence inductionin photosystem II: derivation of analytical expressions in amodel including exciton-Radical pair equilibrium and re-stricted energy transfer between photosynthetic units, Bio-phys. J. 68 (1995) 2474^2492.

[27] Z. Kolber, J. Zehr, P. Falkowski, E¡ects of growth irradi-ance and nitrogen limitation on photosynthetic energy con-version in photosystem II, Plant Physiol. 88 (1988) 923^929.

[28] Z. Kolber, P.G. Falkowski, Use of active £uorescence toestimate phytoplankton photosynthesis in situ, Limnol. Oce-anogr. 38 (1993) 1646^1665.

[29] A.R. Crofts, C.T. Yerkes, A molecular mechanism for qE-quenching, FEBS Lett. 352 (1994) 265^270.

[30] L.L. France, N.E. Geacintov, J. Breton, L. Valkunas, Thedependence of the degrees of sigmoidicities of £uorescenceinduction curves in spinach chloroplasts on the duration ofactinic pulses in Pump-Probe experiments, Biochim. Bio-phys. Acta 1101 (1992) 105^119.

[31] H.W. Trissl, Y. Gao, K. Wulf, Theoretical £uorescence in-duction curves derived from coupled di¡erential equationsdescribing the primary photochemistry of photosystem-IIby an exciton radical pair equilibrium, Biophys. J. 64(1993) 984^998.

[32] A.R. Holzwarth, Is it time to throw away your apparatus forchlorophyll £uorescence induction?, Biophys. J. 64 (1993)1280^1281.

[33] P.G. Falkowski, Z. Kolber, D.C. Mauzerall, A comment onthe call to throw away your £uorescence induction appara-tus, Biophys. J. 66 (1994) 923^925.

[34] U. Schreiber, H. Hormann, C. Neubauer, C. Klughammer,Assessment of photosystem-II photochemical quantum yieldby chlorophyll £uorescence quenching analysis, Aust.J. Plant Physiol. 22 (1995) 209^220.

[35] U. Schreiber, A. Krieger, Two fundamentally di¡erent typesof variable chlorophyll £uorescence in-vivo, FEBS Lett. 397(1996) 131^135.

[36] G. Samson, D. Bruce, Origins of the low-yield of chlorophylla £uorescence induced by single turnover £ash in spinachthylakoids, Biochim. Biophys. Acta 1276 (1996) 147^153.

[37] O. Prasil, Z. Kolber, J.A. Berry, P.G. Falkowski, Cyclicelectron £ow around photosystem-II in-vivo, Photosynth.Res. 48 (1996) 395^410.

[38] H. Kautsky, A. Hirsh, Neue Versuche zur Kohlensauerassi-milation, Naturwissenschaften 48 (1931) 964.

[39] Govindjee, Sixty-three years since Kautsky: chlorophyll a£uorescence, Aust. J. Plant Physiol. 22 (1995) 131^160.

[40] M. Bradbury, N.R. Baker, A quantitative determination ofphotochemical and non-photochemical quenching during theslow phase of the chlorophyll £uorescence induction curve ofbean leaves, Biochim. Biophys. Acta 765 (1984) 275^281.

[41] U. Schreiber, U. Schliwa, W. Bilger, Continuous recordingof photochemical and non-photochemical chlorophyll £uo-rescence quenching with a new type of modulation £uorom-eter, Photosynth. Res. 10 (1986) 51^62.

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106104

[42] R. Delosme, New results about chlorophyll £uorescence invivo, in: G. Forti, A. Melandri (Eds.), Proceedings of the2nd International Congress on Photosynthesis, W. JunkPublishers, The Hague, 1971, pp. 187^195.

[43] D.C. Mauzerall, Light-induced changes in Chlorella and theprimary photoreaction for the production of oxygen, Proc.Natl. Acad. Sci. U.S.A. 69 (1972) 1358^1362.

[44] J.R. Flemming, Chemical Applications of Ultrafast Spectro-scopy, Oxford University Press, New York, 1986.

[45] A. Joliot, E¡ect of low temperature (330 to 360³C) onthe reoxidation of the photosystem II primary electrontransport in the presence and absence of 3(3,4-dichlorphen-yl)-1,1-dimethylurea, Biochim. Biophys. Acta 357 (1974)439^448.

[46] P.G. Falkowski, Y. Fujita, A. Ley, D.C. Mauzerall, Evi-dence for cyclic electron £ow around Photosystem II inChlorella pyrenoidosa, Plant Physiol. 81 (1986) 310^312.

[47] Z.G. Cerovic, M. Bergher, Y. Goulas, S. Tosti, I. Moya,Simultaneous measurement of changes in red and blue £uo-rescence in illuminated isolated chloroplasts and leaf pieces ^the contribution of NADPH to the blue £uorescence signal,Photosynth. Res. 36 (1993) 193^204.

[48] D.M. Kramer, A.R. Crofts, Control and measurement ofphotosynthetic electron transport in vivo, in: N. Baker(Ed.), Photosynthesis and the Environment, Kluwer Aca-demic, Dordrecht, The Netherlands, 1996, pp. 25^66.

[49] M. Trtilek, M. Koblizek, L. Nedbal, Dual-modulation LEDkinetic £uorometer, J. Luminesc. 72-74 (1997) 597.

[50] B. Diner, Dependence of the deactivation reactions of Pho-tosystem II on the redox state of plastoquinone pool A var-ied under anaerobic conditions. Equilibria on the acceptorside of Photosystem II, Biochim. Biophys. Acta 460 (1977)247^258.

[51] M.D. Kramer, G. DiMarco, F. Loreto, Contribution ofplastoquinone quenching to saturation pulse-induced riseof chlorophyll £uorescence in leaves, in: P. Mathis (Ed.),Photosynthesis : From Light to Biosphere, vol. I, KluwerAcademic, Dordrecht, The Netherlands, 1995, pp. 147^150.

[52] G. Deprez, A. Dobek, N.E. Geacintov, G. Paillotin, J. Bre-ton, Probing £orescence induction in chloroplasts on a nano-second time scale utilizing picosecond laser pulse pair, Bio-chim. Biophys. Acta 752 (1983) 444^454.

[53] A. Prasil, Z.S. Kolber, P.G. Falkowski, Measurements ofvariable chlorophyll £uorescence using fast repetition ratetechniques. Modulation of quantum yields by the QB bind-ing site, Photosynth. Res. (1998) submitted.

[54] G. Paillotin, Movement of excitations in the photosyntheticdomains of Photosystem II, J. Theor. Biol. 58 (1976) 237^252.

[55] H.W. Trissl, J. Lavergne, Fluorescence induction from pho-tosystem II: Analytical equations for the yields of photo-chemistry and £uorescence derived from analysis of modelincluding exciton-radical pair equilibrium and restricted en-ergy transfer between photosynthetic units., Aust. J. PlantPhysiol. 22 (1995) 183^193.

[56] J. Breton, N.E. Geacintov, C.E. Swenberg, Quenching of£uorescence by triplet excited states in chloroplasts, Bio-chim. Biophys. Acta 548 (1979) 616^635.

[57] A. Sonnenveld, H. Rademaker, L.N.M. Duysens, Transferand trapping of excitation energy in Photosystem II asstudied by chlorophyll a £uorescence quenching by dinitro-benzene and carotenoid triplet. The matrix model, Biochim.Biophys. Acta 593 (1980) 272^289.

[58] C. Geel, Photosystem II electron £ow as a measure for phy-toplankton gross primary production, Ph.D. Thesis, Depart-ment of Plant Physiology, Wageningen Agricultural Univer-sity, Wageningen, 1997.

[59] J.J. Eaton-Rye, Govindjee, Electron transfer through thequinone acceptor complex of photosystem II in bicarbon-ate-depleted spinach thylakoid membranes as a function ofactinic £ash number and frequency, Biochim. Biophys. Acta935 (1988) 237^247.

[60] I.R. Vassiliev, Z.S. Kolber, D. Mauzerall, V.K. Shukla, K.Wyman, P.G. Falkowski, E¡ects of iron limitation on pho-tosystem II composition and energy trapping in Dunaliellatertiolecta, Plant Physiol. 109 (1995) 963^972.

[61] P. Joliot, A. Joliot, B. Bouges, G. Barbieri, Studies of photo-system II photocenters by comparative measurements of lu-minescence, £uorescence and oxygen emission, PhotochemPhotobiol. 14 (1971) 287^305.

[62] H. Robinson, A.R. Crofts, Kinetics of the changes in theoxidation-reduction state of the acceptors and donors ofphotosystem II in pea thylakoids measured by £ash £uores-cence, in: J. Biggins (Ed.), Progress in Photosynthesis, vol.II, Martinus Nijho¡, Dordrecht, 1987, pp. 429^432.

[63] D.M. Kramer, A.R. Crofts, Demonstration of a highly-sen-sitive portable double-£ash kinetic spectrophotometer formeasurement of electron transfer reactions in intact plants,Photosynth. Res. 23 (1990) 231^240.

[64] M.J. Delrieu, F. Rosengard, Fundamental di¡erences be-tween period-4 oscillations of the oxygen and £uorescenceyield induced by £ash excitation in inside-out thylakoids,Biochim. Biophys. Acta 892 (1987) 163^171.

[65] P. Joliot, A. Joliot, Di¡erent types of quenching involved inphotosystem II centers, Biochim. Biophys. Acta 305 (1973)302^316.

[66] K. Brettel, E. Shlodder, H.T. Witt, Nanosecond reductionkinetics of photooxidized chlorophyll-aII (P-680) in single£ashes as a probe for the electron pathway, H�-release,and charge accumulation in the O2 evolving complex, Bio-chim. Biophys. Acta 766 (1984) 403^415.

[67] C. Neubauer, U. Schreiber, The polyphasic rise of chloro-phyll £uorescence upon onset of strong continuous illumina-tionI. Saturation characteristics and partial control by thephotosystem II acceptor side,, Z Naturforsch c 42 (1987)1246^1254.

[68] R.J. Strasser, A. Srivastava, Govindjee, Polyphasic chloro-phyll-a £uorescence transients in plants and cyanobacteria,Photochem. Photobiol. 61 (1995) 32^42.

[69] A. Srivastava, R.J. Strasser, Govindjee, Polyphasic rise ofchlorophyll-a £uorescence in herbicide-resistant D1 mutants

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106 105

of Chlamydomonas reinhardtii, Photosynth. Res. 43 (1995)131^141.

[70] C. Vernotte, A.L. Etienne, J.M. Briantais, Quenching of thesystem II chlorophyll £uorescence by the plastoquinonepool, Biochim. Biophys. Acta 545 (1979) 519^527.

[71] M.J. Delrieu, F. Rosengard, Events near the reaction centerin O-2 evolving PS-II enriched thylakoid membranes ^ Thepresence of an electric ¢eld during the S-2 state in a popu-lation of centers, Photosynth. Res. 37 (1993) 205^215.

[72] A. Boussac, J.J. Girerd, A.W. Rutherford, Conversion of thespin-state of the manganese complex in photosystem-II in-duced by near-infrared light, Biochemistry 35 (1996) 6984^6989.

[73] V.P. Shinkarev, X. Chunhe, Govindjee, C.A. Wraight, Ki-netics of the oxygen evolution step in plants determinedfrom £ash-induced chlorophyll a £uorescence, Photosynth.Res. 51 (1997) 43^49.

[74] V.P. Shinkarev, C.A. Wraight, Kinetic factors in the bicyclemodel of oxygen evolution by photosystem II, Photosynth.Res. 38 (1993) 315^321.

[75] V.P. Shinkarev, C. Xu, Govindjee, C.A. Wraight, Kinetics ofthe oxygen evolution step in situ estimated from its quench-ing e¡ects on £ash-induced chlorophyll a £uorescence, Bio-phys. J. 66 (1994) A114.

[76] E.D. MacAlister, J. Myers, The time course of photosyn-thesis and £uorescence observed simultaneously, Smithso-nian Institution Publications Miscellaneous Collections 99(1940) 1^37.

[77] W.P. Quick, H. Horton, Studies on the induction of chloro-phyll £uorescence in barley chloroplastsI. Factors a¡ectingthe observations of oscillation in the yield of chlorophyll£uorescence and the rate of oxygen evolution,, Proc. R.Soc. Lond. B 220 (1984) 361^370.

BBABIO 44679 29-9-98

Z.S. Kolber et al. / Biochimica et Biophysica Acta 1367 (1998) 88^106106